Namespace(aa='rand-m9-mstd0.5-inc1', auto_resume=True, batch_size=512, cfg_path='iFormer_t.yaml', clip_grad=None, color_jitter=0.4, crop_pct=None, cutmix=1.0, cutmix_minmax=None, data_path='imagenet', data_set='IMNET', device='cuda', disable_eval=False, dist_backend='nccl', dist_eval=True, dist_on_itp=False, dist_url='env://', distributed=True, drop_path=0, enable_wandb=False, epochs=300, eval=False, eval_data_path=None, finetune='', gpu=0, head_init_scale=1.0, imagenet_default_mean_and_std=True, input_size=224, layer_decay=1.0, layer_scale_init_value=0, local_rank=-1, log_dir=None, lr=0.004, min_lr=1e-06, mixup=0.8, mixup_mode='batch', mixup_prob=1.0, mixup_switch_prob=0.5, model='iFormer_t', model_ema=False, model_ema_decay=0.9999, model_ema_eval=False, model_ema_force_cpu=False, model_key='model|module', model_prefix='', momentum=0.9, nb_classes=1000, num_workers=16, opt='adamw', opt_betas=None, opt_eps=1e-08, output_dir='', pin_mem=True, project='iFormer', rank=0, recount=1, remode='pixel', reprob=0.25, resplit=False, resume='', save_ckpt=True, save_ckpt_freq=1, save_ckpt_num=3, seed=0, smoothing=0.1, start_epoch=0, train_interpolation='bicubic', update_freq=1, use_amp=False, wandb_ckpt=False, warmup_epochs=20, warmup_steps=-1, weight_decay=0.05, weight_decay_end=None, world_size=8)
Transform = 
RandomResizedCropAndInterpolation(size=(224, 224), scale=(0.08, 1.0), ratio=(0.75, 1.3333), interpolation=PIL.Image.BICUBIC)
RandomHorizontalFlip(p=0.5)
<timm.data.auto_augment.RandAugment object at 0x7f806e62f340>
ToTensor()
Normalize(mean=tensor([0.4850, 0.4560, 0.4060]), std=tensor([0.2290, 0.2240, 0.2250]))
<timm.data.random_erasing.RandomErasing object at 0x7f806e62aeb0>
---------------------------
reading from datapath imagenet
Number of the class = 1000
Transform = 
Resize(size=256, interpolation=bicubic, max_size=None, antialias=warn)
CenterCrop(size=(224, 224))
ToTensor()
Normalize(mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225))
---------------------------
reading from datapath imagenet
Number of the class = 1000
Sampler_train = <torch.utils.data.distributed.DistributedSampler object at 0x7f806e62abe0>
Mixup is activated!
Model = iFormer(
  (downsample_layers): ModuleList(
    (0): Sequential(
      (0): Conv2d_BN(
        (c): Conv2d(3, 16, kernel_size=(5, 5), stride=(2, 2), padding=(2, 2), bias=False)
        (bn): BatchNorm2d(16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
      (1): GELU(approximate='none')
      (2): EdgeResidual(
        (conv_exp_bn1): Conv2d_BN(
          (c): Conv2d(16, 64, kernel_size=(5, 5), stride=(2, 2), padding=(2, 2), bias=False)
          (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
        (act): GELU(approximate='none')
        (conv_pwl_bn2): Conv2d_BN(
          (c): Conv2d(64, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
    )
    (1): Sequential(
      (0): Conv2d_BN(
        (c): Conv2d(32, 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
        (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (2): Sequential(
      (0): Conv2d_BN(
        (c): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
        (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (3): Sequential(
      (0): Conv2d_BN(
        (c): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
        (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
  )
  (stages): ModuleList(
    (0): Sequential(
      (0): BasicBlock(
        (block): ConvBlock(
          (token_channel_mixer): Residual(
            (m): Sequential(
              (0): Conv2d_BN(
                (c): Conv2d(32, 32, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=32, bias=False)
                (bn): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(32, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(96, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (1): BasicBlock(
        (block): ConvBlock(
          (token_channel_mixer): Residual(
            (m): Sequential(
              (0): Conv2d_BN(
                (c): Conv2d(32, 32, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=32, bias=False)
                (bn): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(32, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(96, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
    )
    (1): Sequential(
      (0): BasicBlock(
        (block): ConvBlock(
          (token_channel_mixer): Residual(
            (m): Sequential(
              (0): Conv2d_BN(
                (c): Conv2d(64, 64, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=64, bias=False)
                (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(64, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(192, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (1): BasicBlock(
        (block): ConvBlock(
          (token_channel_mixer): Residual(
            (m): Sequential(
              (0): Conv2d_BN(
                (c): Conv2d(64, 64, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=64, bias=False)
                (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(64, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(192, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
    )
    (2): Sequential(
      (0): BasicBlock(
        (block): ConvBlock(
          (token_channel_mixer): Residual(
            (m): Sequential(
              (0): Conv2d_BN(
                (c): Conv2d(128, 128, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=128, bias=False)
                (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(128, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(384, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (1): BasicBlock(
        (block): ConvBlock(
          (token_channel_mixer): Residual(
            (m): Sequential(
              (0): Conv2d_BN(
                (c): Conv2d(128, 128, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=128, bias=False)
                (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(128, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(384, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (2): BasicBlock(
        (block): ConvBlock(
          (token_channel_mixer): Residual(
            (m): Sequential(
              (0): Conv2d_BN(
                (c): Conv2d(128, 128, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=128, bias=False)
                (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(128, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(384, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (3): BasicBlock(
        (block): ConvBlock(
          (token_channel_mixer): Residual(
            (m): Sequential(
              (0): Conv2d_BN(
                (c): Conv2d(128, 128, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=128, bias=False)
                (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(128, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(384, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (4): BasicBlock(
        (block): ConvBlock(
          (token_channel_mixer): Residual(
            (m): Sequential(
              (0): Conv2d_BN(
                (c): Conv2d(128, 128, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=128, bias=False)
                (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(128, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(384, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (5): BasicBlock(
        (block): ConvBlock(
          (token_channel_mixer): Residual(
            (m): Sequential(
              (0): Conv2d_BN(
                (c): Conv2d(128, 128, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=128, bias=False)
                (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(128, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(384, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (6): BasicBlock(
        (block): RepCPE(
          (cpe): Residual(
            (m): Conv2d_BN(
              (c): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=128, bias=False)
              (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            )
          )
        )
      )
      (7): BasicBlock(
        (block): SHMABlock(
          (token_channel_mixer): Residual(
            (m): GAU2dv2(
              (q): Conv2d_BN(
                (c): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (k): Conv2d_BN(
                (c): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (v_gate): Conv2d_BN(
                (c): Conv2d(128, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (gate_act): Sigmoid()
              (attn_drop): Dropout(p=0.0, inplace=False)
              (proj): Conv2d_BN(
                (c): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (8): BasicBlock(
        (block): FFN2d(
          (channel_mixer): Residual(
            (m): Sequential(
              (0): Conv2d_BN(
                (c): Conv2d(128, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): GELU(approximate='none')
              (2): Conv2d_BN(
                (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (9): BasicBlock(
        (block): RepCPE(
          (cpe): Residual(
            (m): Conv2d_BN(
              (c): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=128, bias=False)
              (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            )
          )
        )
      )
      (10): BasicBlock(
        (block): SHMABlock(
          (token_channel_mixer): Residual(
            (m): GAU2dv2(
              (q): Conv2d_BN(
                (c): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (k): Conv2d_BN(
                (c): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (v_gate): Conv2d_BN(
                (c): Conv2d(128, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (gate_act): Sigmoid()
              (attn_drop): Dropout(p=0.0, inplace=False)
              (proj): Conv2d_BN(
                (c): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (11): BasicBlock(
        (block): FFN2d(
          (channel_mixer): Residual(
            (m): Sequential(
              (0): Conv2d_BN(
                (c): Conv2d(128, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): GELU(approximate='none')
              (2): Conv2d_BN(
                (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (12): BasicBlock(
        (block): RepCPE(
          (cpe): Residual(
            (m): Conv2d_BN(
              (c): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=128, bias=False)
              (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            )
          )
        )
      )
      (13): BasicBlock(
        (block): SHMABlock(
          (token_channel_mixer): Residual(
            (m): GAU2dv2(
              (q): Conv2d_BN(
                (c): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (k): Conv2d_BN(
                (c): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (v_gate): Conv2d_BN(
                (c): Conv2d(128, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (gate_act): Sigmoid()
              (attn_drop): Dropout(p=0.0, inplace=False)
              (proj): Conv2d_BN(
                (c): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (14): BasicBlock(
        (block): FFN2d(
          (channel_mixer): Residual(
            (m): Sequential(
              (0): Conv2d_BN(
                (c): Conv2d(128, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): GELU(approximate='none')
              (2): Conv2d_BN(
                (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (15): BasicBlock(
        (block): ConvBlock(
          (token_channel_mixer): Residual(
            (m): Sequential(
              (0): Conv2d_BN(
                (c): Conv2d(128, 128, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=128, bias=False)
                (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(128, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(384, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
    )
    (3): Sequential(
      (0): BasicBlock(
        (block): RepCPE(
          (cpe): Residual(
            (m): Conv2d_BN(
              (c): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=256, bias=False)
              (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            )
          )
        )
      )
      (1): BasicBlock(
        (block): SHMABlock(
          (token_channel_mixer): Residual(
            (m): GAU2dv2(
              (q): Conv2d_BN(
                (c): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (k): Conv2d_BN(
                (c): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (v_gate): Conv2d_BN(
                (c): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (gate_act): Sigmoid()
              (attn_drop): Dropout(p=0.0, inplace=False)
              (proj): Conv2d_BN(
                (c): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (2): BasicBlock(
        (block): FFN2d(
          (channel_mixer): Residual(
            (m): Sequential(
              (0): Conv2d_BN(
                (c): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): GELU(approximate='none')
              (2): Conv2d_BN(
                (c): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (3): BasicBlock(
        (block): RepCPE(
          (cpe): Residual(
            (m): Conv2d_BN(
              (c): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=256, bias=False)
              (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            )
          )
        )
      )
      (4): BasicBlock(
        (block): SHMABlock(
          (token_channel_mixer): Residual(
            (m): GAU2dv2(
              (q): Conv2d_BN(
                (c): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (k): Conv2d_BN(
                (c): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (v_gate): Conv2d_BN(
                (c): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (gate_act): Sigmoid()
              (attn_drop): Dropout(p=0.0, inplace=False)
              (proj): Conv2d_BN(
                (c): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (5): BasicBlock(
        (block): FFN2d(
          (channel_mixer): Residual(
            (m): Sequential(
              (0): Conv2d_BN(
                (c): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): GELU(approximate='none')
              (2): Conv2d_BN(
                (c): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
    )
  )
  (classifier): Classfier(
    (classifier): BN_Linear(
      (bn): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (l): Linear(in_features=256, out_features=1000, bias=True)
    )
  )
)
number of params: 2886456
LR = 0.00400000
Batch size = 4096
Update frequent = 1
Number of training examples = 1281167
Number of training training per epoch = 312
Param groups = {
  "decay": {
    "weight_decay": 0.05,
    "params": [
      "downsample_layers.0.0.c.weight",
      "downsample_layers.0.2.conv_exp_bn1.c.weight",
      "downsample_layers.0.2.conv_pwl_bn2.c.weight",
      "downsample_layers.1.0.c.weight",
      "downsample_layers.2.0.c.weight",
      "downsample_layers.3.0.c.weight",
      "stages.0.0.block.token_channel_mixer.m.0.c.weight",
      "stages.0.0.block.token_channel_mixer.m.1.c.weight",
      "stages.0.0.block.token_channel_mixer.m.3.c.weight",
      "stages.0.1.block.token_channel_mixer.m.0.c.weight",
      "stages.0.1.block.token_channel_mixer.m.1.c.weight",
      "stages.0.1.block.token_channel_mixer.m.3.c.weight",
      "stages.1.0.block.token_channel_mixer.m.0.c.weight",
      "stages.1.0.block.token_channel_mixer.m.1.c.weight",
      "stages.1.0.block.token_channel_mixer.m.3.c.weight",
      "stages.1.1.block.token_channel_mixer.m.0.c.weight",
      "stages.1.1.block.token_channel_mixer.m.1.c.weight",
      "stages.1.1.block.token_channel_mixer.m.3.c.weight",
      "stages.2.0.block.token_channel_mixer.m.0.c.weight",
      "stages.2.0.block.token_channel_mixer.m.1.c.weight",
      "stages.2.0.block.token_channel_mixer.m.3.c.weight",
      "stages.2.1.block.token_channel_mixer.m.0.c.weight",
      "stages.2.1.block.token_channel_mixer.m.1.c.weight",
      "stages.2.1.block.token_channel_mixer.m.3.c.weight",
      "stages.2.2.block.token_channel_mixer.m.0.c.weight",
      "stages.2.2.block.token_channel_mixer.m.1.c.weight",
      "stages.2.2.block.token_channel_mixer.m.3.c.weight",
      "stages.2.3.block.token_channel_mixer.m.0.c.weight",
      "stages.2.3.block.token_channel_mixer.m.1.c.weight",
      "stages.2.3.block.token_channel_mixer.m.3.c.weight",
      "stages.2.4.block.token_channel_mixer.m.0.c.weight",
      "stages.2.4.block.token_channel_mixer.m.1.c.weight",
      "stages.2.4.block.token_channel_mixer.m.3.c.weight",
      "stages.2.5.block.token_channel_mixer.m.0.c.weight",
      "stages.2.5.block.token_channel_mixer.m.1.c.weight",
      "stages.2.5.block.token_channel_mixer.m.3.c.weight",
      "stages.2.6.block.cpe.m.c.weight",
      "stages.2.7.block.token_channel_mixer.m.q.c.weight",
      "stages.2.7.block.token_channel_mixer.m.k.c.weight",
      "stages.2.7.block.token_channel_mixer.m.v_gate.c.weight",
      "stages.2.7.block.token_channel_mixer.m.proj.c.weight",
      "stages.2.8.block.channel_mixer.m.0.c.weight",
      "stages.2.8.block.channel_mixer.m.2.c.weight",
      "stages.2.9.block.cpe.m.c.weight",
      "stages.2.10.block.token_channel_mixer.m.q.c.weight",
      "stages.2.10.block.token_channel_mixer.m.k.c.weight",
      "stages.2.10.block.token_channel_mixer.m.v_gate.c.weight",
      "stages.2.10.block.token_channel_mixer.m.proj.c.weight",
      "stages.2.11.block.channel_mixer.m.0.c.weight",
      "stages.2.11.block.channel_mixer.m.2.c.weight",
      "stages.2.12.block.cpe.m.c.weight",
      "stages.2.13.block.token_channel_mixer.m.q.c.weight",
      "stages.2.13.block.token_channel_mixer.m.k.c.weight",
      "stages.2.13.block.token_channel_mixer.m.v_gate.c.weight",
      "stages.2.13.block.token_channel_mixer.m.proj.c.weight",
      "stages.2.14.block.channel_mixer.m.0.c.weight",
      "stages.2.14.block.channel_mixer.m.2.c.weight",
      "stages.2.15.block.token_channel_mixer.m.0.c.weight",
      "stages.2.15.block.token_channel_mixer.m.1.c.weight",
      "stages.2.15.block.token_channel_mixer.m.3.c.weight",
      "stages.3.0.block.cpe.m.c.weight",
      "stages.3.1.block.token_channel_mixer.m.q.c.weight",
      "stages.3.1.block.token_channel_mixer.m.k.c.weight",
      "stages.3.1.block.token_channel_mixer.m.v_gate.c.weight",
      "stages.3.1.block.token_channel_mixer.m.proj.c.weight",
      "stages.3.2.block.channel_mixer.m.0.c.weight",
      "stages.3.2.block.channel_mixer.m.2.c.weight",
      "stages.3.3.block.cpe.m.c.weight",
      "stages.3.4.block.token_channel_mixer.m.q.c.weight",
      "stages.3.4.block.token_channel_mixer.m.k.c.weight",
      "stages.3.4.block.token_channel_mixer.m.v_gate.c.weight",
      "stages.3.4.block.token_channel_mixer.m.proj.c.weight",
      "stages.3.5.block.channel_mixer.m.0.c.weight",
      "stages.3.5.block.channel_mixer.m.2.c.weight",
      "classifier.classifier.l.weight"
    ],
    "lr_scale": 1.0
  },
  "no_decay": {
    "weight_decay": 0.0,
    "params": [
      "downsample_layers.0.0.bn.weight",
      "downsample_layers.0.0.bn.bias",
      "downsample_layers.0.2.conv_exp_bn1.bn.weight",
      "downsample_layers.0.2.conv_exp_bn1.bn.bias",
      "downsample_layers.0.2.conv_pwl_bn2.bn.weight",
      "downsample_layers.0.2.conv_pwl_bn2.bn.bias",
      "downsample_layers.1.0.bn.weight",
      "downsample_layers.1.0.bn.bias",
      "downsample_layers.2.0.bn.weight",
      "downsample_layers.2.0.bn.bias",
      "downsample_layers.3.0.bn.weight",
      "downsample_layers.3.0.bn.bias",
      "stages.0.0.block.token_channel_mixer.m.0.bn.weight",
      "stages.0.0.block.token_channel_mixer.m.0.bn.bias",
      "stages.0.0.block.token_channel_mixer.m.1.bn.weight",
      "stages.0.0.block.token_channel_mixer.m.1.bn.bias",
      "stages.0.0.block.token_channel_mixer.m.3.bn.weight",
      "stages.0.0.block.token_channel_mixer.m.3.bn.bias",
      "stages.0.1.block.token_channel_mixer.m.0.bn.weight",
      "stages.0.1.block.token_channel_mixer.m.0.bn.bias",
      "stages.0.1.block.token_channel_mixer.m.1.bn.weight",
      "stages.0.1.block.token_channel_mixer.m.1.bn.bias",
      "stages.0.1.block.token_channel_mixer.m.3.bn.weight",
      "stages.0.1.block.token_channel_mixer.m.3.bn.bias",
      "stages.1.0.block.token_channel_mixer.m.0.bn.weight",
      "stages.1.0.block.token_channel_mixer.m.0.bn.bias",
      "stages.1.0.block.token_channel_mixer.m.1.bn.weight",
      "stages.1.0.block.token_channel_mixer.m.1.bn.bias",
      "stages.1.0.block.token_channel_mixer.m.3.bn.weight",
      "stages.1.0.block.token_channel_mixer.m.3.bn.bias",
      "stages.1.1.block.token_channel_mixer.m.0.bn.weight",
      "stages.1.1.block.token_channel_mixer.m.0.bn.bias",
      "stages.1.1.block.token_channel_mixer.m.1.bn.weight",
      "stages.1.1.block.token_channel_mixer.m.1.bn.bias",
      "stages.1.1.block.token_channel_mixer.m.3.bn.weight",
      "stages.1.1.block.token_channel_mixer.m.3.bn.bias",
      "stages.2.0.block.token_channel_mixer.m.0.bn.weight",
      "stages.2.0.block.token_channel_mixer.m.0.bn.bias",
      "stages.2.0.block.token_channel_mixer.m.1.bn.weight",
      "stages.2.0.block.token_channel_mixer.m.1.bn.bias",
      "stages.2.0.block.token_channel_mixer.m.3.bn.weight",
      "stages.2.0.block.token_channel_mixer.m.3.bn.bias",
      "stages.2.1.block.token_channel_mixer.m.0.bn.weight",
      "stages.2.1.block.token_channel_mixer.m.0.bn.bias",
      "stages.2.1.block.token_channel_mixer.m.1.bn.weight",
      "stages.2.1.block.token_channel_mixer.m.1.bn.bias",
      "stages.2.1.block.token_channel_mixer.m.3.bn.weight",
      "stages.2.1.block.token_channel_mixer.m.3.bn.bias",
      "stages.2.2.block.token_channel_mixer.m.0.bn.weight",
      "stages.2.2.block.token_channel_mixer.m.0.bn.bias",
      "stages.2.2.block.token_channel_mixer.m.1.bn.weight",
      "stages.2.2.block.token_channel_mixer.m.1.bn.bias",
      "stages.2.2.block.token_channel_mixer.m.3.bn.weight",
      "stages.2.2.block.token_channel_mixer.m.3.bn.bias",
      "stages.2.3.block.token_channel_mixer.m.0.bn.weight",
      "stages.2.3.block.token_channel_mixer.m.0.bn.bias",
      "stages.2.3.block.token_channel_mixer.m.1.bn.weight",
      "stages.2.3.block.token_channel_mixer.m.1.bn.bias",
      "stages.2.3.block.token_channel_mixer.m.3.bn.weight",
      "stages.2.3.block.token_channel_mixer.m.3.bn.bias",
      "stages.2.4.block.token_channel_mixer.m.0.bn.weight",
      "stages.2.4.block.token_channel_mixer.m.0.bn.bias",
      "stages.2.4.block.token_channel_mixer.m.1.bn.weight",
      "stages.2.4.block.token_channel_mixer.m.1.bn.bias",
      "stages.2.4.block.token_channel_mixer.m.3.bn.weight",
      "stages.2.4.block.token_channel_mixer.m.3.bn.bias",
      "stages.2.5.block.token_channel_mixer.m.0.bn.weight",
      "stages.2.5.block.token_channel_mixer.m.0.bn.bias",
      "stages.2.5.block.token_channel_mixer.m.1.bn.weight",
      "stages.2.5.block.token_channel_mixer.m.1.bn.bias",
      "stages.2.5.block.token_channel_mixer.m.3.bn.weight",
      "stages.2.5.block.token_channel_mixer.m.3.bn.bias",
      "stages.2.6.block.cpe.m.bn.weight",
      "stages.2.6.block.cpe.m.bn.bias",
      "stages.2.7.block.token_channel_mixer.m.q.bn.weight",
      "stages.2.7.block.token_channel_mixer.m.q.bn.bias",
      "stages.2.7.block.token_channel_mixer.m.k.bn.weight",
      "stages.2.7.block.token_channel_mixer.m.k.bn.bias",
      "stages.2.7.block.token_channel_mixer.m.v_gate.bn.weight",
      "stages.2.7.block.token_channel_mixer.m.v_gate.bn.bias",
      "stages.2.7.block.token_channel_mixer.m.proj.bn.weight",
      "stages.2.7.block.token_channel_mixer.m.proj.bn.bias",
      "stages.2.8.block.channel_mixer.m.0.bn.weight",
      "stages.2.8.block.channel_mixer.m.0.bn.bias",
      "stages.2.8.block.channel_mixer.m.2.bn.weight",
      "stages.2.8.block.channel_mixer.m.2.bn.bias",
      "stages.2.9.block.cpe.m.bn.weight",
      "stages.2.9.block.cpe.m.bn.bias",
      "stages.2.10.block.token_channel_mixer.m.q.bn.weight",
      "stages.2.10.block.token_channel_mixer.m.q.bn.bias",
      "stages.2.10.block.token_channel_mixer.m.k.bn.weight",
      "stages.2.10.block.token_channel_mixer.m.k.bn.bias",
      "stages.2.10.block.token_channel_mixer.m.v_gate.bn.weight",
      "stages.2.10.block.token_channel_mixer.m.v_gate.bn.bias",
      "stages.2.10.block.token_channel_mixer.m.proj.bn.weight",
      "stages.2.10.block.token_channel_mixer.m.proj.bn.bias",
      "stages.2.11.block.channel_mixer.m.0.bn.weight",
      "stages.2.11.block.channel_mixer.m.0.bn.bias",
      "stages.2.11.block.channel_mixer.m.2.bn.weight",
      "stages.2.11.block.channel_mixer.m.2.bn.bias",
      "stages.2.12.block.cpe.m.bn.weight",
      "stages.2.12.block.cpe.m.bn.bias",
      "stages.2.13.block.token_channel_mixer.m.q.bn.weight",
      "stages.2.13.block.token_channel_mixer.m.q.bn.bias",
      "stages.2.13.block.token_channel_mixer.m.k.bn.weight",
      "stages.2.13.block.token_channel_mixer.m.k.bn.bias",
      "stages.2.13.block.token_channel_mixer.m.v_gate.bn.weight",
      "stages.2.13.block.token_channel_mixer.m.v_gate.bn.bias",
      "stages.2.13.block.token_channel_mixer.m.proj.bn.weight",
      "stages.2.13.block.token_channel_mixer.m.proj.bn.bias",
      "stages.2.14.block.channel_mixer.m.0.bn.weight",
      "stages.2.14.block.channel_mixer.m.0.bn.bias",
      "stages.2.14.block.channel_mixer.m.2.bn.weight",
      "stages.2.14.block.channel_mixer.m.2.bn.bias",
      "stages.2.15.block.token_channel_mixer.m.0.bn.weight",
      "stages.2.15.block.token_channel_mixer.m.0.bn.bias",
      "stages.2.15.block.token_channel_mixer.m.1.bn.weight",
      "stages.2.15.block.token_channel_mixer.m.1.bn.bias",
      "stages.2.15.block.token_channel_mixer.m.3.bn.weight",
      "stages.2.15.block.token_channel_mixer.m.3.bn.bias",
      "stages.3.0.block.cpe.m.bn.weight",
      "stages.3.0.block.cpe.m.bn.bias",
      "stages.3.1.block.token_channel_mixer.m.q.bn.weight",
      "stages.3.1.block.token_channel_mixer.m.q.bn.bias",
      "stages.3.1.block.token_channel_mixer.m.k.bn.weight",
      "stages.3.1.block.token_channel_mixer.m.k.bn.bias",
      "stages.3.1.block.token_channel_mixer.m.v_gate.bn.weight",
      "stages.3.1.block.token_channel_mixer.m.v_gate.bn.bias",
      "stages.3.1.block.token_channel_mixer.m.proj.bn.weight",
      "stages.3.1.block.token_channel_mixer.m.proj.bn.bias",
      "stages.3.2.block.channel_mixer.m.0.bn.weight",
      "stages.3.2.block.channel_mixer.m.0.bn.bias",
      "stages.3.2.block.channel_mixer.m.2.bn.weight",
      "stages.3.2.block.channel_mixer.m.2.bn.bias",
      "stages.3.3.block.cpe.m.bn.weight",
      "stages.3.3.block.cpe.m.bn.bias",
      "stages.3.4.block.token_channel_mixer.m.q.bn.weight",
      "stages.3.4.block.token_channel_mixer.m.q.bn.bias",
      "stages.3.4.block.token_channel_mixer.m.k.bn.weight",
      "stages.3.4.block.token_channel_mixer.m.k.bn.bias",
      "stages.3.4.block.token_channel_mixer.m.v_gate.bn.weight",
      "stages.3.4.block.token_channel_mixer.m.v_gate.bn.bias",
      "stages.3.4.block.token_channel_mixer.m.proj.bn.weight",
      "stages.3.4.block.token_channel_mixer.m.proj.bn.bias",
      "stages.3.5.block.channel_mixer.m.0.bn.weight",
      "stages.3.5.block.channel_mixer.m.0.bn.bias",
      "stages.3.5.block.channel_mixer.m.2.bn.weight",
      "stages.3.5.block.channel_mixer.m.2.bn.bias",
      "classifier.classifier.bn.weight",
      "classifier.classifier.bn.bias",
      "classifier.classifier.l.bias"
    ],
    "lr_scale": 1.0
  }
}
Use Cosine LR scheduler
Set warmup steps = 6240
Set warmup steps = 0
Max WD = 0.0500000, Min WD = 0.0500000
criterion = SoftTargetCrossEntropy()
Auto resume checkpoint: ./checkpoint-184.pth
Resume checkpoint ./checkpoint-184.pth
With optim & sched!
Start training for 300 epochs
Epoch: [185]  [  0/312]  eta: 3:28:39  lr: 0.001447  min_lr: 0.001447  loss: 4.7776 (4.7776)  weight_decay: 0.0500 (0.0500)  time: 40.1251  data: 21.4799  max mem: 21002
Epoch: [185]  [ 10/312]  eta: 0:20:32  lr: 0.001446  min_lr: 0.001446  loss: 4.3197 (4.2576)  weight_decay: 0.0500 (0.0500)  time: 4.0813  data: 1.9532  max mem: 21002
Epoch: [185]  [ 20/312]  eta: 0:11:02  lr: 0.001446  min_lr: 0.001446  loss: 4.1258 (4.1144)  weight_decay: 0.0500 (0.0500)  time: 0.3774  data: 0.0006  max mem: 21002
Epoch: [185]  [ 30/312]  eta: 0:07:39  lr: 0.001445  min_lr: 0.001445  loss: 4.1043 (4.1046)  weight_decay: 0.0500 (0.0500)  time: 0.2782  data: 0.0006  max mem: 21002
Epoch: [185]  [ 40/312]  eta: 0:06:49  lr: 0.001444  min_lr: 0.001444  loss: 3.9541 (4.0304)  weight_decay: 0.0500 (0.0500)  time: 0.7042  data: 0.3828  max mem: 21002
Epoch: [185]  [ 50/312]  eta: 0:06:26  lr: 0.001444  min_lr: 0.001444  loss: 3.8703 (4.0131)  weight_decay: 0.0500 (0.0500)  time: 1.2351  data: 0.8635  max mem: 21002
Epoch: [185]  [ 60/312]  eta: 0:05:35  lr: 0.001443  min_lr: 0.001443  loss: 3.9302 (3.9807)  weight_decay: 0.0500 (0.0500)  time: 0.9770  data: 0.5809  max mem: 21002
Epoch: [185]  [ 70/312]  eta: 0:05:24  lr: 0.001442  min_lr: 0.001442  loss: 3.9200 (3.9395)  weight_decay: 0.0500 (0.0500)  time: 0.9999  data: 0.6010  max mem: 21002
Epoch: [185]  [ 80/312]  eta: 0:04:42  lr: 0.001441  min_lr: 0.001441  loss: 3.8287 (3.9173)  weight_decay: 0.0500 (0.0500)  time: 0.8643  data: 0.5015  max mem: 21002
Epoch: [185]  [ 90/312]  eta: 0:04:31  lr: 0.001441  min_lr: 0.001441  loss: 3.9179 (3.9136)  weight_decay: 0.0500 (0.0500)  time: 0.8141  data: 0.4236  max mem: 21002
Epoch: [185]  [100/312]  eta: 0:04:17  lr: 0.001440  min_lr: 0.001440  loss: 4.0469 (3.9254)  weight_decay: 0.0500 (0.0500)  time: 1.2128  data: 0.7309  max mem: 21002
Epoch: [185]  [110/312]  eta: 0:03:53  lr: 0.001439  min_lr: 0.001439  loss: 4.0469 (3.9285)  weight_decay: 0.0500 (0.0500)  time: 0.8341  data: 0.4102  max mem: 21002
Epoch: [185]  [120/312]  eta: 0:03:44  lr: 0.001439  min_lr: 0.001439  loss: 3.9577 (3.9325)  weight_decay: 0.0500 (0.0500)  time: 0.9251  data: 0.5630  max mem: 21002
Epoch: [185]  [130/312]  eta: 0:03:31  lr: 0.001438  min_lr: 0.001438  loss: 3.9283 (3.9191)  weight_decay: 0.0500 (0.0500)  time: 1.1961  data: 0.7978  max mem: 21002
Epoch: [185]  [140/312]  eta: 0:03:13  lr: 0.001437  min_lr: 0.001437  loss: 3.7867 (3.9053)  weight_decay: 0.0500 (0.0500)  time: 0.8616  data: 0.4782  max mem: 21002
Epoch: [185]  [150/312]  eta: 0:03:05  lr: 0.001437  min_lr: 0.001437  loss: 3.7464 (3.8951)  weight_decay: 0.0500 (0.0500)  time: 1.0321  data: 0.5708  max mem: 21002
Epoch: [185]  [160/312]  eta: 0:02:45  lr: 0.001436  min_lr: 0.001436  loss: 3.9646 (3.9011)  weight_decay: 0.0500 (0.0500)  time: 0.8511  data: 0.4303  max mem: 21002
Epoch: [185]  [170/312]  eta: 0:02:37  lr: 0.001435  min_lr: 0.001435  loss: 3.9700 (3.8935)  weight_decay: 0.0500 (0.0500)  time: 0.8449  data: 0.4511  max mem: 21002
Epoch: [185]  [180/312]  eta: 0:02:25  lr: 0.001435  min_lr: 0.001435  loss: 3.7780 (3.8790)  weight_decay: 0.0500 (0.0500)  time: 1.2164  data: 0.7146  max mem: 21002
Epoch: [185]  [190/312]  eta: 0:02:10  lr: 0.001434  min_lr: 0.001434  loss: 3.7688 (3.8758)  weight_decay: 0.0500 (0.0500)  time: 0.7711  data: 0.3790  max mem: 21002
Epoch: [185]  [200/312]  eta: 0:02:01  lr: 0.001433  min_lr: 0.001433  loss: 3.6115 (3.8608)  weight_decay: 0.0500 (0.0500)  time: 0.8918  data: 0.5042  max mem: 21002
Epoch: [185]  [210/312]  eta: 0:01:50  lr: 0.001432  min_lr: 0.001432  loss: 3.8248 (3.8548)  weight_decay: 0.0500 (0.0500)  time: 1.1313  data: 0.6378  max mem: 21002
Epoch: [185]  [220/312]  eta: 0:01:37  lr: 0.001432  min_lr: 0.001432  loss: 3.9037 (3.8480)  weight_decay: 0.0500 (0.0500)  time: 0.8406  data: 0.4300  max mem: 21002
Epoch: [185]  [230/312]  eta: 0:01:28  lr: 0.001431  min_lr: 0.001431  loss: 3.8900 (3.8443)  weight_decay: 0.0500 (0.0500)  time: 1.0816  data: 0.6117  max mem: 21002
Epoch: [185]  [240/312]  eta: 0:01:15  lr: 0.001430  min_lr: 0.001430  loss: 3.9297 (3.8436)  weight_decay: 0.0500 (0.0500)  time: 0.8735  data: 0.4309  max mem: 21002
Epoch: [185]  [250/312]  eta: 0:01:05  lr: 0.001430  min_lr: 0.001430  loss: 3.7176 (3.8357)  weight_decay: 0.0500 (0.0500)  time: 0.8808  data: 0.4207  max mem: 21002
Epoch: [185]  [260/312]  eta: 0:00:55  lr: 0.001429  min_lr: 0.001429  loss: 3.6910 (3.8268)  weight_decay: 0.0500 (0.0500)  time: 1.3171  data: 0.6734  max mem: 21002
Epoch: [185]  [270/312]  eta: 0:00:43  lr: 0.001428  min_lr: 0.001428  loss: 3.6211 (3.8215)  weight_decay: 0.0500 (0.0500)  time: 0.8236  data: 0.3524  max mem: 21002
Epoch: [185]  [280/312]  eta: 0:00:33  lr: 0.001428  min_lr: 0.001428  loss: 4.0236 (3.8273)  weight_decay: 0.0500 (0.0500)  time: 0.9296  data: 0.3895  max mem: 21002
Epoch: [185]  [290/312]  eta: 0:00:23  lr: 0.001427  min_lr: 0.001427  loss: 4.0236 (3.8202)  weight_decay: 0.0500 (0.0500)  time: 1.1625  data: 0.3881  max mem: 21002
Epoch: [185]  [300/312]  eta: 0:00:12  lr: 0.001426  min_lr: 0.001426  loss: 3.4791 (3.8090)  weight_decay: 0.0500 (0.0500)  time: 0.8002  data: 0.2368  max mem: 21002
Epoch: [185]  [310/312]  eta: 0:00:02  lr: 0.001426  min_lr: 0.001426  loss: 3.6150 (3.8066)  weight_decay: 0.0500 (0.0500)  time: 0.6982  data: 0.2672  max mem: 21002
Epoch: [185]  [311/312]  eta: 0:00:01  lr: 0.001426  min_lr: 0.001426  loss: 3.6150 (3.8073)  weight_decay: 0.0500 (0.0500)  time: 0.6775  data: 0.2672  max mem: 21002
Epoch: [185] Total time: 0:05:22 (1.0321 s / it)
Averaged stats: lr: 0.001426  min_lr: 0.001426  loss: 3.6150 (3.8315)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:03:17  loss: 1.4889 (1.4889)  acc1: 70.3125 (70.3125)  acc5: 90.3646 (90.3646)  time: 21.9968  data: 17.9279  max mem: 21002
Test:  [8/9]  eta: 0:00:02  loss: 1.9074 (1.8183)  acc1: 63.2812 (63.7280)  acc5: 85.1562 (85.2160)  time: 2.6415  data: 1.9921  max mem: 21002
Test: Total time: 0:00:23 (2.6533 s / it)
* Acc@1 63.080 Acc@5 85.666 loss 1.814
Accuracy of the model on the 50000 test images: 63.1%
Max accuracy: 63.08%
Epoch: [186]  [  0/312]  eta: 1:26:41  lr: 0.001425  min_lr: 0.001425  loss: 4.6332 (4.6332)  weight_decay: 0.0500 (0.0500)  time: 16.6699  data: 11.4587  max mem: 21002
Epoch: [186]  [ 10/312]  eta: 0:10:00  lr: 0.001425  min_lr: 0.001425  loss: 4.2259 (4.2220)  weight_decay: 0.0500 (0.0500)  time: 1.9887  data: 1.0421  max mem: 21002
Epoch: [186]  [ 20/312]  eta: 0:06:17  lr: 0.001424  min_lr: 0.001424  loss: 4.1311 (4.0367)  weight_decay: 0.0500 (0.0500)  time: 0.5222  data: 0.0006  max mem: 21002
Epoch: [186]  [ 30/312]  eta: 0:04:52  lr: 0.001423  min_lr: 0.001423  loss: 4.0918 (3.9926)  weight_decay: 0.0500 (0.0500)  time: 0.5145  data: 0.0007  max mem: 21002
Epoch: [186]  [ 40/312]  eta: 0:04:06  lr: 0.001423  min_lr: 0.001423  loss: 4.0401 (3.9831)  weight_decay: 0.0500 (0.0500)  time: 0.4990  data: 0.0007  max mem: 21002
Epoch: [186]  [ 50/312]  eta: 0:03:34  lr: 0.001422  min_lr: 0.001422  loss: 3.8079 (3.9437)  weight_decay: 0.0500 (0.0500)  time: 0.4822  data: 0.0009  max mem: 21002
Epoch: [186]  [ 60/312]  eta: 0:03:10  lr: 0.001421  min_lr: 0.001421  loss: 3.8025 (3.9034)  weight_decay: 0.0500 (0.0500)  time: 0.4567  data: 0.0010  max mem: 21002
Epoch: [186]  [ 70/312]  eta: 0:02:52  lr: 0.001421  min_lr: 0.001421  loss: 3.8025 (3.8713)  weight_decay: 0.0500 (0.0500)  time: 0.4410  data: 0.0011  max mem: 21002
Epoch: [186]  [ 80/312]  eta: 0:02:37  lr: 0.001420  min_lr: 0.001420  loss: 3.7597 (3.8502)  weight_decay: 0.0500 (0.0500)  time: 0.4466  data: 0.0010  max mem: 21002
Epoch: [186]  [ 90/312]  eta: 0:02:25  lr: 0.001419  min_lr: 0.001419  loss: 4.0322 (3.8707)  weight_decay: 0.0500 (0.0500)  time: 0.4486  data: 0.0008  max mem: 21002
Epoch: [186]  [100/312]  eta: 0:02:14  lr: 0.001419  min_lr: 0.001419  loss: 3.9802 (3.8827)  weight_decay: 0.0500 (0.0500)  time: 0.4435  data: 0.0016  max mem: 21002
Epoch: [186]  [110/312]  eta: 0:02:04  lr: 0.001418  min_lr: 0.001418  loss: 3.9384 (3.8691)  weight_decay: 0.0500 (0.0500)  time: 0.4428  data: 0.0028  max mem: 21002
Epoch: [186]  [120/312]  eta: 0:01:55  lr: 0.001417  min_lr: 0.001417  loss: 4.0920 (3.9020)  weight_decay: 0.0500 (0.0500)  time: 0.4441  data: 0.0020  max mem: 21002
Epoch: [186]  [130/312]  eta: 0:01:48  lr: 0.001416  min_lr: 0.001416  loss: 4.1042 (3.9061)  weight_decay: 0.0500 (0.0500)  time: 0.4662  data: 0.0239  max mem: 21002
Epoch: [186]  [140/312]  eta: 0:01:41  lr: 0.001416  min_lr: 0.001416  loss: 3.8015 (3.8862)  weight_decay: 0.0500 (0.0500)  time: 0.5135  data: 0.0946  max mem: 21002
Epoch: [186]  [150/312]  eta: 0:01:34  lr: 0.001415  min_lr: 0.001415  loss: 3.6452 (3.8728)  weight_decay: 0.0500 (0.0500)  time: 0.5211  data: 0.1418  max mem: 21002
Epoch: [186]  [160/312]  eta: 0:01:29  lr: 0.001414  min_lr: 0.001414  loss: 3.5607 (3.8486)  weight_decay: 0.0500 (0.0500)  time: 0.5732  data: 0.1631  max mem: 21002
Epoch: [186]  [170/312]  eta: 0:01:21  lr: 0.001414  min_lr: 0.001414  loss: 3.7257 (3.8534)  weight_decay: 0.0500 (0.0500)  time: 0.5125  data: 0.1352  max mem: 21002
Epoch: [186]  [180/312]  eta: 0:01:16  lr: 0.001413  min_lr: 0.001413  loss: 3.9088 (3.8550)  weight_decay: 0.0500 (0.0500)  time: 0.5313  data: 0.1064  max mem: 21002
Epoch: [186]  [190/312]  eta: 0:01:10  lr: 0.001412  min_lr: 0.001412  loss: 3.8104 (3.8486)  weight_decay: 0.0500 (0.0500)  time: 0.6336  data: 0.0822  max mem: 21002
Epoch: [186]  [200/312]  eta: 0:01:04  lr: 0.001412  min_lr: 0.001412  loss: 3.9454 (3.8691)  weight_decay: 0.0500 (0.0500)  time: 0.4885  data: 0.0713  max mem: 21002
Epoch: [186]  [210/312]  eta: 0:00:59  lr: 0.001411  min_lr: 0.001411  loss: 3.9158 (3.8575)  weight_decay: 0.0500 (0.0500)  time: 0.5529  data: 0.1342  max mem: 21002
Epoch: [186]  [220/312]  eta: 0:00:52  lr: 0.001410  min_lr: 0.001410  loss: 3.8995 (3.8535)  weight_decay: 0.0500 (0.0500)  time: 0.5961  data: 0.0960  max mem: 21002
Epoch: [186]  [230/312]  eta: 0:00:47  lr: 0.001410  min_lr: 0.001410  loss: 4.0158 (3.8465)  weight_decay: 0.0500 (0.0500)  time: 0.5277  data: 0.0810  max mem: 21002
Epoch: [186]  [240/312]  eta: 0:00:41  lr: 0.001409  min_lr: 0.001409  loss: 4.0709 (3.8547)  weight_decay: 0.0500 (0.0500)  time: 0.6248  data: 0.1249  max mem: 21002
Epoch: [186]  [250/312]  eta: 0:00:35  lr: 0.001408  min_lr: 0.001408  loss: 4.1024 (3.8606)  weight_decay: 0.0500 (0.0500)  time: 0.4797  data: 0.0588  max mem: 21002
Epoch: [186]  [260/312]  eta: 0:00:29  lr: 0.001408  min_lr: 0.001408  loss: 4.0443 (3.8614)  weight_decay: 0.0500 (0.0500)  time: 0.5083  data: 0.0483  max mem: 21002
Epoch: [186]  [270/312]  eta: 0:00:24  lr: 0.001407  min_lr: 0.001407  loss: 4.0904 (3.8701)  weight_decay: 0.0500 (0.0500)  time: 0.6736  data: 0.0484  max mem: 21002
Epoch: [186]  [280/312]  eta: 0:00:18  lr: 0.001406  min_lr: 0.001406  loss: 4.1406 (3.8694)  weight_decay: 0.0500 (0.0500)  time: 0.5016  data: 0.0506  max mem: 21002
Epoch: [186]  [290/312]  eta: 0:00:12  lr: 0.001406  min_lr: 0.001406  loss: 3.8005 (3.8631)  weight_decay: 0.0500 (0.0500)  time: 0.5189  data: 0.0844  max mem: 21002
Epoch: [186]  [300/312]  eta: 0:00:06  lr: 0.001405  min_lr: 0.001405  loss: 3.8368 (3.8659)  weight_decay: 0.0500 (0.0500)  time: 0.4800  data: 0.0343  max mem: 21002
Epoch: [186]  [310/312]  eta: 0:00:01  lr: 0.001404  min_lr: 0.001404  loss: 3.9907 (3.8746)  weight_decay: 0.0500 (0.0500)  time: 0.3121  data: 0.0002  max mem: 21002
Epoch: [186]  [311/312]  eta: 0:00:00  lr: 0.001404  min_lr: 0.001404  loss: 4.1389 (3.8757)  weight_decay: 0.0500 (0.0500)  time: 0.3119  data: 0.0002  max mem: 21002
Epoch: [186] Total time: 0:02:53 (0.5575 s / it)
Averaged stats: lr: 0.001404  min_lr: 0.001404  loss: 4.1389 (3.8393)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:23  loss: 1.6118 (1.6118)  acc1: 69.6615 (69.6615)  acc5: 89.8438 (89.8438)  time: 9.2424  data: 9.1235  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 2.1096 (1.9515)  acc1: 58.7240 (61.9520)  acc5: 84.5052 (85.0880)  time: 1.1154  data: 1.0175  max mem: 21002
Test: Total time: 0:00:10 (1.1301 s / it)
* Acc@1 62.208 Acc@5 85.350 loss 1.968
Accuracy of the model on the 50000 test images: 62.2%
Max accuracy: 63.08%
Epoch: [187]  [  0/312]  eta: 1:02:12  lr: 0.001404  min_lr: 0.001404  loss: 4.1415 (4.1415)  weight_decay: 0.0500 (0.0500)  time: 11.9629  data: 10.0203  max mem: 21002
Epoch: [187]  [ 10/312]  eta: 0:07:49  lr: 0.001403  min_lr: 0.001403  loss: 3.7426 (3.8852)  weight_decay: 0.0500 (0.0500)  time: 1.5546  data: 1.0413  max mem: 21002
Epoch: [187]  [ 20/312]  eta: 0:05:08  lr: 0.001403  min_lr: 0.001403  loss: 3.7066 (3.7140)  weight_decay: 0.0500 (0.0500)  time: 0.5100  data: 0.1489  max mem: 21002
Epoch: [187]  [ 30/312]  eta: 0:03:47  lr: 0.001402  min_lr: 0.001402  loss: 3.7508 (3.7773)  weight_decay: 0.0500 (0.0500)  time: 0.3979  data: 0.0775  max mem: 21002
Epoch: [187]  [ 40/312]  eta: 0:03:25  lr: 0.001401  min_lr: 0.001401  loss: 3.6976 (3.7060)  weight_decay: 0.0500 (0.0500)  time: 0.4371  data: 0.1506  max mem: 21002
Epoch: [187]  [ 50/312]  eta: 0:03:13  lr: 0.001401  min_lr: 0.001401  loss: 4.0243 (3.8134)  weight_decay: 0.0500 (0.0500)  time: 0.6318  data: 0.3473  max mem: 21002
Epoch: [187]  [ 60/312]  eta: 0:02:47  lr: 0.001400  min_lr: 0.001400  loss: 4.0504 (3.8140)  weight_decay: 0.0500 (0.0500)  time: 0.4828  data: 0.1973  max mem: 21002
Epoch: [187]  [ 70/312]  eta: 0:02:42  lr: 0.001399  min_lr: 0.001399  loss: 3.6309 (3.7916)  weight_decay: 0.0500 (0.0500)  time: 0.5045  data: 0.2183  max mem: 21002
Epoch: [187]  [ 80/312]  eta: 0:02:35  lr: 0.001399  min_lr: 0.001399  loss: 3.6131 (3.7790)  weight_decay: 0.0500 (0.0500)  time: 0.6841  data: 0.3965  max mem: 21002
Epoch: [187]  [ 90/312]  eta: 0:02:19  lr: 0.001398  min_lr: 0.001398  loss: 3.4239 (3.7468)  weight_decay: 0.0500 (0.0500)  time: 0.4648  data: 0.1797  max mem: 21002
Epoch: [187]  [100/312]  eta: 0:02:14  lr: 0.001397  min_lr: 0.001397  loss: 3.7394 (3.7540)  weight_decay: 0.0500 (0.0500)  time: 0.4840  data: 0.1873  max mem: 21002
Epoch: [187]  [110/312]  eta: 0:02:01  lr: 0.001396  min_lr: 0.001396  loss: 3.9885 (3.7726)  weight_decay: 0.0500 (0.0500)  time: 0.4860  data: 0.1864  max mem: 21002
Epoch: [187]  [120/312]  eta: 0:01:56  lr: 0.001396  min_lr: 0.001396  loss: 4.0255 (3.7903)  weight_decay: 0.0500 (0.0500)  time: 0.4618  data: 0.1725  max mem: 21002
Epoch: [187]  [130/312]  eta: 0:01:51  lr: 0.001395  min_lr: 0.001395  loss: 4.0255 (3.7809)  weight_decay: 0.0500 (0.0500)  time: 0.6589  data: 0.3712  max mem: 21002
Epoch: [187]  [140/312]  eta: 0:01:41  lr: 0.001394  min_lr: 0.001394  loss: 4.0443 (3.8058)  weight_decay: 0.0500 (0.0500)  time: 0.4944  data: 0.1994  max mem: 21002
Epoch: [187]  [150/312]  eta: 0:01:36  lr: 0.001394  min_lr: 0.001394  loss: 4.1434 (3.8161)  weight_decay: 0.0500 (0.0500)  time: 0.4997  data: 0.1798  max mem: 21002
Epoch: [187]  [160/312]  eta: 0:01:30  lr: 0.001393  min_lr: 0.001393  loss: 4.0065 (3.8116)  weight_decay: 0.0500 (0.0500)  time: 0.6642  data: 0.3537  max mem: 21002
Epoch: [187]  [170/312]  eta: 0:01:22  lr: 0.001392  min_lr: 0.001392  loss: 3.8901 (3.8165)  weight_decay: 0.0500 (0.0500)  time: 0.4881  data: 0.1745  max mem: 21002
Epoch: [187]  [180/312]  eta: 0:01:17  lr: 0.001392  min_lr: 0.001392  loss: 3.8901 (3.8091)  weight_decay: 0.0500 (0.0500)  time: 0.5140  data: 0.1920  max mem: 21002
Epoch: [187]  [190/312]  eta: 0:01:09  lr: 0.001391  min_lr: 0.001391  loss: 3.4707 (3.7880)  weight_decay: 0.0500 (0.0500)  time: 0.4864  data: 0.1919  max mem: 21002
Epoch: [187]  [200/312]  eta: 0:01:04  lr: 0.001390  min_lr: 0.001390  loss: 3.3687 (3.7881)  weight_decay: 0.0500 (0.0500)  time: 0.4917  data: 0.2062  max mem: 21002
Epoch: [187]  [210/312]  eta: 0:00:59  lr: 0.001390  min_lr: 0.001390  loss: 3.8016 (3.7775)  weight_decay: 0.0500 (0.0500)  time: 0.6833  data: 0.3937  max mem: 21002
Epoch: [187]  [220/312]  eta: 0:00:52  lr: 0.001389  min_lr: 0.001389  loss: 3.4292 (3.7720)  weight_decay: 0.0500 (0.0500)  time: 0.4791  data: 0.1882  max mem: 21002
Epoch: [187]  [230/312]  eta: 0:00:47  lr: 0.001388  min_lr: 0.001388  loss: 3.9478 (3.7801)  weight_decay: 0.0500 (0.0500)  time: 0.4921  data: 0.2045  max mem: 21002
Epoch: [187]  [240/312]  eta: 0:00:41  lr: 0.001388  min_lr: 0.001388  loss: 3.9204 (3.7781)  weight_decay: 0.0500 (0.0500)  time: 0.6763  data: 0.3900  max mem: 21002
Epoch: [187]  [250/312]  eta: 0:00:35  lr: 0.001387  min_lr: 0.001387  loss: 3.9003 (3.7847)  weight_decay: 0.0500 (0.0500)  time: 0.4725  data: 0.1862  max mem: 21002
Epoch: [187]  [260/312]  eta: 0:00:29  lr: 0.001386  min_lr: 0.001386  loss: 3.8030 (3.7776)  weight_decay: 0.0500 (0.0500)  time: 0.4835  data: 0.1971  max mem: 21002
Epoch: [187]  [270/312]  eta: 0:00:23  lr: 0.001385  min_lr: 0.001385  loss: 3.7837 (3.7791)  weight_decay: 0.0500 (0.0500)  time: 0.4864  data: 0.1971  max mem: 21002
Epoch: [187]  [280/312]  eta: 0:00:18  lr: 0.001385  min_lr: 0.001385  loss: 4.0251 (3.7828)  weight_decay: 0.0500 (0.0500)  time: 0.4744  data: 0.1818  max mem: 21002
Epoch: [187]  [290/312]  eta: 0:00:12  lr: 0.001384  min_lr: 0.001384  loss: 4.0251 (3.7819)  weight_decay: 0.0500 (0.0500)  time: 0.5931  data: 0.3048  max mem: 21002
Epoch: [187]  [300/312]  eta: 0:00:06  lr: 0.001383  min_lr: 0.001383  loss: 4.1027 (3.7945)  weight_decay: 0.0500 (0.0500)  time: 0.4040  data: 0.1233  max mem: 21002
Epoch: [187]  [310/312]  eta: 0:00:01  lr: 0.001383  min_lr: 0.001383  loss: 4.1027 (3.7997)  weight_decay: 0.0500 (0.0500)  time: 0.2773  data: 0.0001  max mem: 21002
Epoch: [187]  [311/312]  eta: 0:00:00  lr: 0.001383  min_lr: 0.001383  loss: 4.0232 (3.7965)  weight_decay: 0.0500 (0.0500)  time: 0.2772  data: 0.0001  max mem: 21002
Epoch: [187] Total time: 0:02:51 (0.5481 s / it)
Averaged stats: lr: 0.001383  min_lr: 0.001383  loss: 4.0232 (3.8194)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:23  loss: 1.7346 (1.7346)  acc1: 69.0104 (69.0104)  acc5: 91.2760 (91.2760)  time: 9.3126  data: 9.1906  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 2.1113 (2.0165)  acc1: 62.1094 (62.6400)  acc5: 85.8073 (85.5520)  time: 1.1290  data: 1.0261  max mem: 21002
Test: Total time: 0:00:10 (1.1463 s / it)
* Acc@1 62.462 Acc@5 85.344 loss 2.040
Accuracy of the model on the 50000 test images: 62.5%
Max accuracy: 63.08%
Epoch: [188]  [  0/312]  eta: 1:01:05  lr: 0.001383  min_lr: 0.001383  loss: 4.2814 (4.2814)  weight_decay: 0.0500 (0.0500)  time: 11.7491  data: 10.9142  max mem: 21002
Epoch: [188]  [ 10/312]  eta: 0:07:35  lr: 0.001382  min_lr: 0.001382  loss: 3.8155 (3.6153)  weight_decay: 0.0500 (0.0500)  time: 1.5074  data: 0.9928  max mem: 21002
Epoch: [188]  [ 20/312]  eta: 0:05:09  lr: 0.001381  min_lr: 0.001381  loss: 3.7004 (3.5595)  weight_decay: 0.0500 (0.0500)  time: 0.5265  data: 0.1310  max mem: 21002
Epoch: [188]  [ 30/312]  eta: 0:03:48  lr: 0.001381  min_lr: 0.001381  loss: 3.7375 (3.6529)  weight_decay: 0.0500 (0.0500)  time: 0.4294  data: 0.1310  max mem: 21002
Epoch: [188]  [ 40/312]  eta: 0:03:28  lr: 0.001380  min_lr: 0.001380  loss: 3.9737 (3.7171)  weight_decay: 0.0500 (0.0500)  time: 0.4567  data: 0.1065  max mem: 21002
Epoch: [188]  [ 50/312]  eta: 0:03:16  lr: 0.001379  min_lr: 0.001379  loss: 3.9737 (3.7365)  weight_decay: 0.0500 (0.0500)  time: 0.6552  data: 0.2441  max mem: 21002
Epoch: [188]  [ 60/312]  eta: 0:02:49  lr: 0.001379  min_lr: 0.001379  loss: 3.9066 (3.7338)  weight_decay: 0.0500 (0.0500)  time: 0.4858  data: 0.1382  max mem: 21002
Epoch: [188]  [ 70/312]  eta: 0:02:43  lr: 0.001378  min_lr: 0.001378  loss: 3.8160 (3.7431)  weight_decay: 0.0500 (0.0500)  time: 0.4862  data: 0.1161  max mem: 21002
Epoch: [188]  [ 80/312]  eta: 0:02:34  lr: 0.001377  min_lr: 0.001377  loss: 4.0151 (3.7777)  weight_decay: 0.0500 (0.0500)  time: 0.6475  data: 0.2306  max mem: 21002
Epoch: [188]  [ 90/312]  eta: 0:02:20  lr: 0.001376  min_lr: 0.001376  loss: 4.0068 (3.7508)  weight_decay: 0.0500 (0.0500)  time: 0.4705  data: 0.1153  max mem: 21002
Epoch: [188]  [100/312]  eta: 0:02:15  lr: 0.001376  min_lr: 0.001376  loss: 3.6596 (3.7638)  weight_decay: 0.0500 (0.0500)  time: 0.5126  data: 0.1240  max mem: 21002
Epoch: [188]  [110/312]  eta: 0:02:02  lr: 0.001375  min_lr: 0.001375  loss: 4.0125 (3.7800)  weight_decay: 0.0500 (0.0500)  time: 0.4887  data: 0.1244  max mem: 21002
Epoch: [188]  [120/312]  eta: 0:01:57  lr: 0.001374  min_lr: 0.001374  loss: 4.1170 (3.7970)  weight_decay: 0.0500 (0.0500)  time: 0.4985  data: 0.1167  max mem: 21002
Epoch: [188]  [130/312]  eta: 0:01:53  lr: 0.001374  min_lr: 0.001374  loss: 4.0867 (3.8041)  weight_decay: 0.0500 (0.0500)  time: 0.7184  data: 0.2387  max mem: 21002
Epoch: [188]  [140/312]  eta: 0:01:42  lr: 0.001373  min_lr: 0.001373  loss: 3.5936 (3.7876)  weight_decay: 0.0500 (0.0500)  time: 0.5061  data: 0.1232  max mem: 21002
Epoch: [188]  [150/312]  eta: 0:01:38  lr: 0.001372  min_lr: 0.001372  loss: 3.7967 (3.7879)  weight_decay: 0.0500 (0.0500)  time: 0.5127  data: 0.1170  max mem: 21002
Epoch: [188]  [160/312]  eta: 0:01:31  lr: 0.001372  min_lr: 0.001372  loss: 3.9163 (3.7905)  weight_decay: 0.0500 (0.0500)  time: 0.6132  data: 0.1177  max mem: 21002
Epoch: [188]  [170/312]  eta: 0:01:24  lr: 0.001371  min_lr: 0.001371  loss: 3.9163 (3.7894)  weight_decay: 0.0500 (0.0500)  time: 0.4977  data: 0.0887  max mem: 21002
Epoch: [188]  [180/312]  eta: 0:01:18  lr: 0.001370  min_lr: 0.001370  loss: 4.0368 (3.7982)  weight_decay: 0.0500 (0.0500)  time: 0.5365  data: 0.1430  max mem: 21002
Epoch: [188]  [190/312]  eta: 0:01:10  lr: 0.001370  min_lr: 0.001370  loss: 4.0933 (3.8048)  weight_decay: 0.0500 (0.0500)  time: 0.4277  data: 0.0559  max mem: 21002
Epoch: [188]  [200/312]  eta: 0:01:05  lr: 0.001369  min_lr: 0.001369  loss: 3.9175 (3.8030)  weight_decay: 0.0500 (0.0500)  time: 0.4818  data: 0.1137  max mem: 21002
Epoch: [188]  [210/312]  eta: 0:01:00  lr: 0.001368  min_lr: 0.001368  loss: 3.9379 (3.8058)  weight_decay: 0.0500 (0.0500)  time: 0.7212  data: 0.2499  max mem: 21002
Epoch: [188]  [220/312]  eta: 0:00:53  lr: 0.001368  min_lr: 0.001368  loss: 3.9808 (3.8081)  weight_decay: 0.0500 (0.0500)  time: 0.5274  data: 0.1369  max mem: 21002
Epoch: [188]  [230/312]  eta: 0:00:47  lr: 0.001367  min_lr: 0.001367  loss: 3.8129 (3.7962)  weight_decay: 0.0500 (0.0500)  time: 0.4974  data: 0.0896  max mem: 21002
Epoch: [188]  [240/312]  eta: 0:00:41  lr: 0.001366  min_lr: 0.001366  loss: 4.0012 (3.8147)  weight_decay: 0.0500 (0.0500)  time: 0.5794  data: 0.1028  max mem: 21002
Epoch: [188]  [250/312]  eta: 0:00:35  lr: 0.001366  min_lr: 0.001366  loss: 4.2199 (3.8290)  weight_decay: 0.0500 (0.0500)  time: 0.4528  data: 0.0722  max mem: 21002
Epoch: [188]  [260/312]  eta: 0:00:30  lr: 0.001365  min_lr: 0.001365  loss: 3.9395 (3.8268)  weight_decay: 0.0500 (0.0500)  time: 0.5911  data: 0.1773  max mem: 21002
Epoch: [188]  [270/312]  eta: 0:00:23  lr: 0.001364  min_lr: 0.001364  loss: 3.8643 (3.8356)  weight_decay: 0.0500 (0.0500)  time: 0.5079  data: 0.1189  max mem: 21002
Epoch: [188]  [280/312]  eta: 0:00:18  lr: 0.001364  min_lr: 0.001364  loss: 4.0621 (3.8380)  weight_decay: 0.0500 (0.0500)  time: 0.5012  data: 0.0881  max mem: 21002
Epoch: [188]  [290/312]  eta: 0:00:12  lr: 0.001363  min_lr: 0.001363  loss: 3.8255 (3.8293)  weight_decay: 0.0500 (0.0500)  time: 0.6001  data: 0.1287  max mem: 21002
Epoch: [188]  [300/312]  eta: 0:00:06  lr: 0.001362  min_lr: 0.001362  loss: 3.6892 (3.8251)  weight_decay: 0.0500 (0.0500)  time: 0.3822  data: 0.0410  max mem: 21002
Epoch: [188]  [310/312]  eta: 0:00:01  lr: 0.001361  min_lr: 0.001361  loss: 3.9144 (3.8286)  weight_decay: 0.0500 (0.0500)  time: 0.2770  data: 0.0001  max mem: 21002
Epoch: [188]  [311/312]  eta: 0:00:00  lr: 0.001361  min_lr: 0.001361  loss: 3.9144 (3.8270)  weight_decay: 0.0500 (0.0500)  time: 0.2769  data: 0.0001  max mem: 21002
Epoch: [188] Total time: 0:02:52 (0.5542 s / it)
Averaged stats: lr: 0.001361  min_lr: 0.001361  loss: 3.9144 (3.8198)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:24  loss: 1.5904 (1.5904)  acc1: 72.2656 (72.2656)  acc5: 90.3646 (90.3646)  time: 9.4206  data: 9.3020  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.8049 (1.8085)  acc1: 62.3698 (62.0960)  acc5: 85.2865 (84.6400)  time: 1.1332  data: 1.0337  max mem: 21002
Test: Total time: 0:00:10 (1.1434 s / it)
* Acc@1 62.020 Acc@5 84.650 loss 1.810
Accuracy of the model on the 50000 test images: 62.0%
Max accuracy: 63.08%
Epoch: [189]  [  0/312]  eta: 1:02:35  lr: 0.001361  min_lr: 0.001361  loss: 2.6549 (2.6549)  weight_decay: 0.0500 (0.0500)  time: 12.0359  data: 10.3591  max mem: 21002
Epoch: [189]  [ 10/312]  eta: 0:08:43  lr: 0.001361  min_lr: 0.001361  loss: 3.9349 (3.7189)  weight_decay: 0.0500 (0.0500)  time: 1.7331  data: 1.1040  max mem: 21002
Epoch: [189]  [ 20/312]  eta: 0:05:11  lr: 0.001360  min_lr: 0.001360  loss: 3.9512 (3.8681)  weight_decay: 0.0500 (0.0500)  time: 0.5193  data: 0.0993  max mem: 21002
Epoch: [189]  [ 30/312]  eta: 0:03:50  lr: 0.001359  min_lr: 0.001359  loss: 4.0370 (3.8947)  weight_decay: 0.0500 (0.0500)  time: 0.3124  data: 0.0105  max mem: 21002
Epoch: [189]  [ 40/312]  eta: 0:03:16  lr: 0.001359  min_lr: 0.001359  loss: 4.0656 (3.9249)  weight_decay: 0.0500 (0.0500)  time: 0.3589  data: 0.0014  max mem: 21002
Epoch: [189]  [ 50/312]  eta: 0:03:03  lr: 0.001358  min_lr: 0.001358  loss: 3.9855 (3.9124)  weight_decay: 0.0500 (0.0500)  time: 0.5186  data: 0.0234  max mem: 21002
Epoch: [189]  [ 60/312]  eta: 0:02:41  lr: 0.001357  min_lr: 0.001357  loss: 3.8250 (3.8827)  weight_decay: 0.0500 (0.0500)  time: 0.4702  data: 0.0227  max mem: 21002
Epoch: [189]  [ 70/312]  eta: 0:02:35  lr: 0.001357  min_lr: 0.001357  loss: 3.6131 (3.8310)  weight_decay: 0.0500 (0.0500)  time: 0.4901  data: 0.0617  max mem: 21002
Epoch: [189]  [ 80/312]  eta: 0:02:26  lr: 0.001356  min_lr: 0.001356  loss: 3.6131 (3.8235)  weight_decay: 0.0500 (0.0500)  time: 0.6123  data: 0.1413  max mem: 21002
Epoch: [189]  [ 90/312]  eta: 0:02:16  lr: 0.001355  min_lr: 0.001355  loss: 4.0239 (3.8261)  weight_decay: 0.0500 (0.0500)  time: 0.5244  data: 0.1205  max mem: 21002
Epoch: [189]  [100/312]  eta: 0:02:09  lr: 0.001355  min_lr: 0.001355  loss: 3.8915 (3.8301)  weight_decay: 0.0500 (0.0500)  time: 0.5216  data: 0.1021  max mem: 21002
Epoch: [189]  [110/312]  eta: 0:01:58  lr: 0.001354  min_lr: 0.001354  loss: 3.9432 (3.8481)  weight_decay: 0.0500 (0.0500)  time: 0.4584  data: 0.0921  max mem: 21002
Epoch: [189]  [120/312]  eta: 0:01:54  lr: 0.001353  min_lr: 0.001353  loss: 4.0814 (3.8608)  weight_decay: 0.0500 (0.0500)  time: 0.5326  data: 0.1430  max mem: 21002
Epoch: [189]  [130/312]  eta: 0:01:49  lr: 0.001352  min_lr: 0.001352  loss: 4.0683 (3.8680)  weight_decay: 0.0500 (0.0500)  time: 0.6674  data: 0.1959  max mem: 21002
Epoch: [189]  [140/312]  eta: 0:01:40  lr: 0.001352  min_lr: 0.001352  loss: 4.0683 (3.8723)  weight_decay: 0.0500 (0.0500)  time: 0.4862  data: 0.1013  max mem: 21002
Epoch: [189]  [150/312]  eta: 0:01:34  lr: 0.001351  min_lr: 0.001351  loss: 4.0054 (3.8704)  weight_decay: 0.0500 (0.0500)  time: 0.4925  data: 0.0965  max mem: 21002
Epoch: [189]  [160/312]  eta: 0:01:28  lr: 0.001350  min_lr: 0.001350  loss: 3.9960 (3.8826)  weight_decay: 0.0500 (0.0500)  time: 0.5780  data: 0.0913  max mem: 21002
Epoch: [189]  [170/312]  eta: 0:01:22  lr: 0.001350  min_lr: 0.001350  loss: 3.9607 (3.8761)  weight_decay: 0.0500 (0.0500)  time: 0.5391  data: 0.0622  max mem: 21002
Epoch: [189]  [180/312]  eta: 0:01:16  lr: 0.001349  min_lr: 0.001349  loss: 3.8411 (3.8686)  weight_decay: 0.0500 (0.0500)  time: 0.5813  data: 0.0909  max mem: 21002
Epoch: [189]  [190/312]  eta: 0:01:09  lr: 0.001348  min_lr: 0.001348  loss: 3.8411 (3.8666)  weight_decay: 0.0500 (0.0500)  time: 0.5095  data: 0.0664  max mem: 21002
Epoch: [189]  [200/312]  eta: 0:01:04  lr: 0.001348  min_lr: 0.001348  loss: 4.0771 (3.8720)  weight_decay: 0.0500 (0.0500)  time: 0.5301  data: 0.0847  max mem: 21002
Epoch: [189]  [210/312]  eta: 0:00:58  lr: 0.001347  min_lr: 0.001347  loss: 4.1394 (3.8749)  weight_decay: 0.0500 (0.0500)  time: 0.6138  data: 0.0722  max mem: 21002
Epoch: [189]  [220/312]  eta: 0:00:52  lr: 0.001346  min_lr: 0.001346  loss: 4.0132 (3.8781)  weight_decay: 0.0500 (0.0500)  time: 0.5489  data: 0.0370  max mem: 21002
Epoch: [189]  [230/312]  eta: 0:00:46  lr: 0.001346  min_lr: 0.001346  loss: 3.9125 (3.8618)  weight_decay: 0.0500 (0.0500)  time: 0.5239  data: 0.0389  max mem: 21002
Epoch: [189]  [240/312]  eta: 0:00:41  lr: 0.001345  min_lr: 0.001345  loss: 3.4789 (3.8444)  weight_decay: 0.0500 (0.0500)  time: 0.5281  data: 0.0399  max mem: 21002
Epoch: [189]  [250/312]  eta: 0:00:35  lr: 0.001344  min_lr: 0.001344  loss: 3.4442 (3.8321)  weight_decay: 0.0500 (0.0500)  time: 0.5913  data: 0.0798  max mem: 21002
Epoch: [189]  [260/312]  eta: 0:00:29  lr: 0.001344  min_lr: 0.001344  loss: 3.9300 (3.8406)  weight_decay: 0.0500 (0.0500)  time: 0.5681  data: 0.0697  max mem: 21002
Epoch: [189]  [270/312]  eta: 0:00:23  lr: 0.001343  min_lr: 0.001343  loss: 3.9939 (3.8428)  weight_decay: 0.0500 (0.0500)  time: 0.5069  data: 0.0613  max mem: 21002
Epoch: [189]  [280/312]  eta: 0:00:18  lr: 0.001342  min_lr: 0.001342  loss: 3.9149 (3.8445)  weight_decay: 0.0500 (0.0500)  time: 0.5411  data: 0.0927  max mem: 21002
Epoch: [189]  [290/312]  eta: 0:00:12  lr: 0.001342  min_lr: 0.001342  loss: 3.8324 (3.8412)  weight_decay: 0.0500 (0.0500)  time: 0.5265  data: 0.0625  max mem: 21002
Epoch: [189]  [300/312]  eta: 0:00:06  lr: 0.001341  min_lr: 0.001341  loss: 3.6636 (3.8334)  weight_decay: 0.0500 (0.0500)  time: 0.4333  data: 0.0305  max mem: 21002
Epoch: [189]  [310/312]  eta: 0:00:01  lr: 0.001340  min_lr: 0.001340  loss: 3.8734 (3.8343)  weight_decay: 0.0500 (0.0500)  time: 0.3250  data: 0.0184  max mem: 21002
Epoch: [189]  [311/312]  eta: 0:00:00  lr: 0.001340  min_lr: 0.001340  loss: 3.7811 (3.8317)  weight_decay: 0.0500 (0.0500)  time: 0.3249  data: 0.0184  max mem: 21002
Epoch: [189] Total time: 0:02:52 (0.5528 s / it)
Averaged stats: lr: 0.001340  min_lr: 0.001340  loss: 3.7811 (3.8132)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:23  loss: 1.7435 (1.7435)  acc1: 69.6615 (69.6615)  acc5: 90.3646 (90.3646)  time: 9.3195  data: 9.2007  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 2.1031 (1.9596)  acc1: 60.4167 (61.2960)  acc5: 85.8073 (84.6560)  time: 1.1276  data: 1.0224  max mem: 21002
Test: Total time: 0:00:10 (1.1500 s / it)
* Acc@1 61.702 Acc@5 84.888 loss 1.957
Accuracy of the model on the 50000 test images: 61.7%
Max accuracy: 63.08%
Epoch: [190]  [  0/312]  eta: 1:07:47  lr: 0.001340  min_lr: 0.001340  loss: 2.5269 (2.5269)  weight_decay: 0.0500 (0.0500)  time: 13.0374  data: 12.7440  max mem: 21002
Epoch: [190]  [ 10/312]  eta: 0:08:24  lr: 0.001339  min_lr: 0.001339  loss: 3.5821 (3.4284)  weight_decay: 0.0500 (0.0500)  time: 1.6712  data: 1.1603  max mem: 21002
Epoch: [190]  [ 20/312]  eta: 0:05:10  lr: 0.001339  min_lr: 0.001339  loss: 3.5888 (3.5853)  weight_decay: 0.0500 (0.0500)  time: 0.4664  data: 0.0581  max mem: 21002
Epoch: [190]  [ 30/312]  eta: 0:03:49  lr: 0.001338  min_lr: 0.001338  loss: 4.0021 (3.7416)  weight_decay: 0.0500 (0.0500)  time: 0.3445  data: 0.0576  max mem: 21002
Epoch: [190]  [ 40/312]  eta: 0:03:18  lr: 0.001337  min_lr: 0.001337  loss: 4.0966 (3.8253)  weight_decay: 0.0500 (0.0500)  time: 0.3803  data: 0.0928  max mem: 21002
Epoch: [190]  [ 50/312]  eta: 0:03:08  lr: 0.001337  min_lr: 0.001337  loss: 4.0169 (3.8439)  weight_decay: 0.0500 (0.0500)  time: 0.5662  data: 0.2819  max mem: 21002
Epoch: [190]  [ 60/312]  eta: 0:02:43  lr: 0.001336  min_lr: 0.001336  loss: 3.9110 (3.8311)  weight_decay: 0.0500 (0.0500)  time: 0.4775  data: 0.1944  max mem: 21002
Epoch: [190]  [ 70/312]  eta: 0:02:38  lr: 0.001335  min_lr: 0.001335  loss: 3.9848 (3.8451)  weight_decay: 0.0500 (0.0500)  time: 0.4954  data: 0.2104  max mem: 21002
Epoch: [190]  [ 80/312]  eta: 0:02:31  lr: 0.001335  min_lr: 0.001335  loss: 3.9910 (3.8423)  weight_decay: 0.0500 (0.0500)  time: 0.6718  data: 0.3844  max mem: 21002
Epoch: [190]  [ 90/312]  eta: 0:02:16  lr: 0.001334  min_lr: 0.001334  loss: 3.7193 (3.7826)  weight_decay: 0.0500 (0.0500)  time: 0.4727  data: 0.1794  max mem: 21002
Epoch: [190]  [100/312]  eta: 0:02:11  lr: 0.001333  min_lr: 0.001333  loss: 3.2921 (3.7506)  weight_decay: 0.0500 (0.0500)  time: 0.4951  data: 0.2007  max mem: 21002
Epoch: [190]  [110/312]  eta: 0:02:00  lr: 0.001333  min_lr: 0.001333  loss: 3.5748 (3.7572)  weight_decay: 0.0500 (0.0500)  time: 0.5031  data: 0.2007  max mem: 21002
Epoch: [190]  [120/312]  eta: 0:01:56  lr: 0.001332  min_lr: 0.001332  loss: 4.0577 (3.7634)  weight_decay: 0.0500 (0.0500)  time: 0.5184  data: 0.2182  max mem: 21002
Epoch: [190]  [130/312]  eta: 0:01:50  lr: 0.001331  min_lr: 0.001331  loss: 3.9018 (3.7638)  weight_decay: 0.0500 (0.0500)  time: 0.6779  data: 0.3915  max mem: 21002
Epoch: [190]  [140/312]  eta: 0:01:41  lr: 0.001331  min_lr: 0.001331  loss: 3.9675 (3.7940)  weight_decay: 0.0500 (0.0500)  time: 0.4882  data: 0.1745  max mem: 21002
Epoch: [190]  [150/312]  eta: 0:01:35  lr: 0.001330  min_lr: 0.001330  loss: 4.0010 (3.7907)  weight_decay: 0.0500 (0.0500)  time: 0.4925  data: 0.1791  max mem: 21002
Epoch: [190]  [160/312]  eta: 0:01:30  lr: 0.001329  min_lr: 0.001329  loss: 3.6965 (3.7820)  weight_decay: 0.0500 (0.0500)  time: 0.6481  data: 0.3479  max mem: 21002
Epoch: [190]  [170/312]  eta: 0:01:22  lr: 0.001329  min_lr: 0.001329  loss: 4.0714 (3.7993)  weight_decay: 0.0500 (0.0500)  time: 0.5006  data: 0.1699  max mem: 21002
Epoch: [190]  [180/312]  eta: 0:01:17  lr: 0.001328  min_lr: 0.001328  loss: 4.0680 (3.7957)  weight_decay: 0.0500 (0.0500)  time: 0.4995  data: 0.1837  max mem: 21002
Epoch: [190]  [190/312]  eta: 0:01:09  lr: 0.001327  min_lr: 0.001327  loss: 3.5809 (3.7884)  weight_decay: 0.0500 (0.0500)  time: 0.4893  data: 0.1838  max mem: 21002
Epoch: [190]  [200/312]  eta: 0:01:04  lr: 0.001327  min_lr: 0.001327  loss: 3.4890 (3.7892)  weight_decay: 0.0500 (0.0500)  time: 0.5053  data: 0.1889  max mem: 21002
Epoch: [190]  [210/312]  eta: 0:00:59  lr: 0.001326  min_lr: 0.001326  loss: 3.9044 (3.7952)  weight_decay: 0.0500 (0.0500)  time: 0.6502  data: 0.3543  max mem: 21002
Epoch: [190]  [220/312]  eta: 0:00:52  lr: 0.001325  min_lr: 0.001325  loss: 4.0180 (3.7982)  weight_decay: 0.0500 (0.0500)  time: 0.4824  data: 0.1665  max mem: 21002
Epoch: [190]  [230/312]  eta: 0:00:46  lr: 0.001325  min_lr: 0.001325  loss: 3.8735 (3.7926)  weight_decay: 0.0500 (0.0500)  time: 0.5055  data: 0.1918  max mem: 21002
Epoch: [190]  [240/312]  eta: 0:00:41  lr: 0.001324  min_lr: 0.001324  loss: 4.0118 (3.8037)  weight_decay: 0.0500 (0.0500)  time: 0.6732  data: 0.3771  max mem: 21002
Epoch: [190]  [250/312]  eta: 0:00:35  lr: 0.001323  min_lr: 0.001323  loss: 4.0118 (3.7992)  weight_decay: 0.0500 (0.0500)  time: 0.5193  data: 0.1864  max mem: 21002
Epoch: [190]  [260/312]  eta: 0:00:29  lr: 0.001322  min_lr: 0.001322  loss: 3.8818 (3.8041)  weight_decay: 0.0500 (0.0500)  time: 0.4939  data: 0.1711  max mem: 21002
Epoch: [190]  [270/312]  eta: 0:00:23  lr: 0.001322  min_lr: 0.001322  loss: 3.9297 (3.8108)  weight_decay: 0.0500 (0.0500)  time: 0.4945  data: 0.1712  max mem: 21002
Epoch: [190]  [280/312]  eta: 0:00:18  lr: 0.001321  min_lr: 0.001321  loss: 3.9180 (3.8073)  weight_decay: 0.0500 (0.0500)  time: 0.5256  data: 0.1923  max mem: 21002
Epoch: [190]  [290/312]  eta: 0:00:12  lr: 0.001320  min_lr: 0.001320  loss: 3.6733 (3.8007)  weight_decay: 0.0500 (0.0500)  time: 0.6234  data: 0.3233  max mem: 21002
Epoch: [190]  [300/312]  eta: 0:00:06  lr: 0.001320  min_lr: 0.001320  loss: 3.8414 (3.8014)  weight_decay: 0.0500 (0.0500)  time: 0.4160  data: 0.1317  max mem: 21002
Epoch: [190]  [310/312]  eta: 0:00:01  lr: 0.001319  min_lr: 0.001319  loss: 3.8414 (3.7946)  weight_decay: 0.0500 (0.0500)  time: 0.2775  data: 0.0002  max mem: 21002
Epoch: [190]  [311/312]  eta: 0:00:00  lr: 0.001319  min_lr: 0.001319  loss: 3.8414 (3.7932)  weight_decay: 0.0500 (0.0500)  time: 0.2775  data: 0.0002  max mem: 21002
Epoch: [190] Total time: 0:02:52 (0.5517 s / it)
Averaged stats: lr: 0.001319  min_lr: 0.001319  loss: 3.8414 (3.8157)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:22  loss: 1.4860 (1.4860)  acc1: 71.7448 (71.7448)  acc5: 88.6719 (88.6719)  time: 9.1168  data: 9.0006  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.6496 (1.5995)  acc1: 64.3229 (65.2320)  acc5: 86.3281 (86.9920)  time: 1.1010  data: 1.0001  max mem: 21002
Test: Total time: 0:00:10 (1.1242 s / it)
* Acc@1 65.106 Acc@5 87.048 loss 1.602
Accuracy of the model on the 50000 test images: 65.1%
Max accuracy: 65.11%
Epoch: [191]  [  0/312]  eta: 1:00:36  lr: 0.001319  min_lr: 0.001319  loss: 4.7072 (4.7072)  weight_decay: 0.0500 (0.0500)  time: 11.6570  data: 10.0700  max mem: 21002
Epoch: [191]  [ 10/312]  eta: 0:07:54  lr: 0.001318  min_lr: 0.001318  loss: 4.0483 (4.0673)  weight_decay: 0.0500 (0.0500)  time: 1.5698  data: 1.0015  max mem: 21002
Epoch: [191]  [ 20/312]  eta: 0:04:54  lr: 0.001318  min_lr: 0.001318  loss: 3.9375 (3.9358)  weight_decay: 0.0500 (0.0500)  time: 0.4775  data: 0.0908  max mem: 21002
Epoch: [191]  [ 30/312]  eta: 0:03:39  lr: 0.001317  min_lr: 0.001317  loss: 3.8413 (3.8750)  weight_decay: 0.0500 (0.0500)  time: 0.3432  data: 0.0482  max mem: 21002
Epoch: [191]  [ 40/312]  eta: 0:03:18  lr: 0.001316  min_lr: 0.001316  loss: 3.8413 (3.8509)  weight_decay: 0.0500 (0.0500)  time: 0.4325  data: 0.0982  max mem: 21002
Epoch: [191]  [ 50/312]  eta: 0:03:03  lr: 0.001316  min_lr: 0.001316  loss: 3.6076 (3.7595)  weight_decay: 0.0500 (0.0500)  time: 0.5759  data: 0.2091  max mem: 21002
Epoch: [191]  [ 60/312]  eta: 0:02:44  lr: 0.001315  min_lr: 0.001315  loss: 3.4332 (3.7363)  weight_decay: 0.0500 (0.0500)  time: 0.4967  data: 0.1831  max mem: 21002
Epoch: [191]  [ 70/312]  eta: 0:02:37  lr: 0.001314  min_lr: 0.001314  loss: 3.8348 (3.7418)  weight_decay: 0.0500 (0.0500)  time: 0.5255  data: 0.2376  max mem: 21002
Epoch: [191]  [ 80/312]  eta: 0:02:33  lr: 0.001314  min_lr: 0.001314  loss: 3.7276 (3.7136)  weight_decay: 0.0500 (0.0500)  time: 0.6836  data: 0.3923  max mem: 21002
Epoch: [191]  [ 90/312]  eta: 0:02:17  lr: 0.001313  min_lr: 0.001313  loss: 3.8742 (3.7269)  weight_decay: 0.0500 (0.0500)  time: 0.5079  data: 0.2229  max mem: 21002
Epoch: [191]  [100/312]  eta: 0:02:13  lr: 0.001312  min_lr: 0.001312  loss: 3.8884 (3.7075)  weight_decay: 0.0500 (0.0500)  time: 0.5014  data: 0.2177  max mem: 21002
Epoch: [191]  [110/312]  eta: 0:02:00  lr: 0.001312  min_lr: 0.001312  loss: 3.7323 (3.7046)  weight_decay: 0.0500 (0.0500)  time: 0.5019  data: 0.2173  max mem: 21002
Epoch: [191]  [120/312]  eta: 0:01:55  lr: 0.001311  min_lr: 0.001311  loss: 3.7549 (3.7036)  weight_decay: 0.0500 (0.0500)  time: 0.4758  data: 0.1909  max mem: 21002
Epoch: [191]  [130/312]  eta: 0:01:49  lr: 0.001310  min_lr: 0.001310  loss: 3.8552 (3.7072)  weight_decay: 0.0500 (0.0500)  time: 0.6235  data: 0.3364  max mem: 21002
Epoch: [191]  [140/312]  eta: 0:01:39  lr: 0.001310  min_lr: 0.001310  loss: 3.9293 (3.7321)  weight_decay: 0.0500 (0.0500)  time: 0.4342  data: 0.1462  max mem: 21002
Epoch: [191]  [150/312]  eta: 0:01:35  lr: 0.001309  min_lr: 0.001309  loss: 3.9601 (3.7391)  weight_decay: 0.0500 (0.0500)  time: 0.5164  data: 0.2294  max mem: 21002
Epoch: [191]  [160/312]  eta: 0:01:29  lr: 0.001308  min_lr: 0.001308  loss: 4.0674 (3.7556)  weight_decay: 0.0500 (0.0500)  time: 0.6724  data: 0.3867  max mem: 21002
Epoch: [191]  [170/312]  eta: 0:01:21  lr: 0.001307  min_lr: 0.001307  loss: 4.0663 (3.7615)  weight_decay: 0.0500 (0.0500)  time: 0.4686  data: 0.1872  max mem: 21002
Epoch: [191]  [180/312]  eta: 0:01:16  lr: 0.001307  min_lr: 0.001307  loss: 3.8075 (3.7598)  weight_decay: 0.0500 (0.0500)  time: 0.5090  data: 0.2140  max mem: 21002
Epoch: [191]  [190/312]  eta: 0:01:09  lr: 0.001306  min_lr: 0.001306  loss: 3.5206 (3.7470)  weight_decay: 0.0500 (0.0500)  time: 0.5170  data: 0.2012  max mem: 21002
Epoch: [191]  [200/312]  eta: 0:01:04  lr: 0.001305  min_lr: 0.001305  loss: 3.8651 (3.7627)  weight_decay: 0.0500 (0.0500)  time: 0.5010  data: 0.1644  max mem: 21002
Epoch: [191]  [210/312]  eta: 0:00:58  lr: 0.001305  min_lr: 0.001305  loss: 3.9345 (3.7572)  weight_decay: 0.0500 (0.0500)  time: 0.6574  data: 0.2618  max mem: 21002
Epoch: [191]  [220/312]  eta: 0:00:52  lr: 0.001304  min_lr: 0.001304  loss: 3.9345 (3.7733)  weight_decay: 0.0500 (0.0500)  time: 0.5625  data: 0.1380  max mem: 21002
Epoch: [191]  [230/312]  eta: 0:00:46  lr: 0.001303  min_lr: 0.001303  loss: 4.0720 (3.7712)  weight_decay: 0.0500 (0.0500)  time: 0.5125  data: 0.1288  max mem: 21002
Epoch: [191]  [240/312]  eta: 0:00:41  lr: 0.001303  min_lr: 0.001303  loss: 3.7359 (3.7723)  weight_decay: 0.0500 (0.0500)  time: 0.5602  data: 0.1423  max mem: 21002
Epoch: [191]  [250/312]  eta: 0:00:35  lr: 0.001302  min_lr: 0.001302  loss: 3.8765 (3.7756)  weight_decay: 0.0500 (0.0500)  time: 0.5521  data: 0.1237  max mem: 21002
Epoch: [191]  [260/312]  eta: 0:00:29  lr: 0.001301  min_lr: 0.001301  loss: 4.0639 (3.7858)  weight_decay: 0.0500 (0.0500)  time: 0.5676  data: 0.2049  max mem: 21002
Epoch: [191]  [270/312]  eta: 0:00:23  lr: 0.001301  min_lr: 0.001301  loss: 4.1689 (3.7906)  weight_decay: 0.0500 (0.0500)  time: 0.4836  data: 0.1191  max mem: 21002
Epoch: [191]  [280/312]  eta: 0:00:18  lr: 0.001300  min_lr: 0.001300  loss: 4.0748 (3.7919)  weight_decay: 0.0500 (0.0500)  time: 0.4856  data: 0.1094  max mem: 21002
Epoch: [191]  [290/312]  eta: 0:00:12  lr: 0.001299  min_lr: 0.001299  loss: 4.0648 (3.7974)  weight_decay: 0.0500 (0.0500)  time: 0.5728  data: 0.1687  max mem: 21002
Epoch: [191]  [300/312]  eta: 0:00:06  lr: 0.001299  min_lr: 0.001299  loss: 4.0102 (3.7989)  weight_decay: 0.0500 (0.0500)  time: 0.4180  data: 0.0598  max mem: 21002
Epoch: [191]  [310/312]  eta: 0:00:01  lr: 0.001298  min_lr: 0.001298  loss: 3.9497 (3.7990)  weight_decay: 0.0500 (0.0500)  time: 0.2772  data: 0.0001  max mem: 21002
Epoch: [191]  [311/312]  eta: 0:00:00  lr: 0.001298  min_lr: 0.001298  loss: 3.9347 (3.7968)  weight_decay: 0.0500 (0.0500)  time: 0.2772  data: 0.0001  max mem: 21002
Epoch: [191] Total time: 0:02:51 (0.5494 s / it)
Averaged stats: lr: 0.001298  min_lr: 0.001298  loss: 3.9347 (3.8026)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:24  loss: 1.3352 (1.3352)  acc1: 72.3958 (72.3958)  acc5: 91.0156 (91.0156)  time: 9.3880  data: 9.2693  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.7483 (1.6087)  acc1: 64.0625 (65.5200)  acc5: 85.1562 (86.5440)  time: 1.1280  data: 1.0300  max mem: 21002
Test: Total time: 0:00:10 (1.1427 s / it)
* Acc@1 64.778 Acc@5 86.934 loss 1.620
Accuracy of the model on the 50000 test images: 64.8%
Max accuracy: 65.11%
Epoch: [192]  [  0/312]  eta: 1:02:22  lr: 0.001298  min_lr: 0.001298  loss: 4.2022 (4.2022)  weight_decay: 0.0500 (0.0500)  time: 11.9942  data: 9.2953  max mem: 21002
Epoch: [192]  [ 10/312]  eta: 0:07:39  lr: 0.001297  min_lr: 0.001297  loss: 4.1816 (4.1044)  weight_decay: 0.0500 (0.0500)  time: 1.5203  data: 1.0126  max mem: 21002
Epoch: [192]  [ 20/312]  eta: 0:04:53  lr: 0.001297  min_lr: 0.001297  loss: 3.8714 (3.8251)  weight_decay: 0.0500 (0.0500)  time: 0.4548  data: 0.1528  max mem: 21002
Epoch: [192]  [ 30/312]  eta: 0:03:37  lr: 0.001296  min_lr: 0.001296  loss: 3.7059 (3.8173)  weight_decay: 0.0500 (0.0500)  time: 0.3611  data: 0.0610  max mem: 21002
Epoch: [192]  [ 40/312]  eta: 0:03:18  lr: 0.001295  min_lr: 0.001295  loss: 3.5190 (3.7165)  weight_decay: 0.0500 (0.0500)  time: 0.4410  data: 0.1346  max mem: 21002
Epoch: [192]  [ 50/312]  eta: 0:03:08  lr: 0.001295  min_lr: 0.001295  loss: 3.5190 (3.7139)  weight_decay: 0.0500 (0.0500)  time: 0.6353  data: 0.3046  max mem: 21002
Epoch: [192]  [ 60/312]  eta: 0:02:43  lr: 0.001294  min_lr: 0.001294  loss: 3.6690 (3.7088)  weight_decay: 0.0500 (0.0500)  time: 0.4801  data: 0.1706  max mem: 21002
Epoch: [192]  [ 70/312]  eta: 0:02:39  lr: 0.001293  min_lr: 0.001293  loss: 3.5038 (3.7141)  weight_decay: 0.0500 (0.0500)  time: 0.5001  data: 0.1918  max mem: 21002
Epoch: [192]  [ 80/312]  eta: 0:02:31  lr: 0.001293  min_lr: 0.001293  loss: 4.0026 (3.7471)  weight_decay: 0.0500 (0.0500)  time: 0.6629  data: 0.3232  max mem: 21002
Epoch: [192]  [ 90/312]  eta: 0:02:16  lr: 0.001292  min_lr: 0.001292  loss: 4.0470 (3.7702)  weight_decay: 0.0500 (0.0500)  time: 0.4606  data: 0.1323  max mem: 21002
Epoch: [192]  [100/312]  eta: 0:02:12  lr: 0.001291  min_lr: 0.001291  loss: 3.8334 (3.7527)  weight_decay: 0.0500 (0.0500)  time: 0.5207  data: 0.1233  max mem: 21002
Epoch: [192]  [110/312]  eta: 0:02:00  lr: 0.001291  min_lr: 0.001291  loss: 3.8334 (3.7555)  weight_decay: 0.0500 (0.0500)  time: 0.5122  data: 0.1281  max mem: 21002
Epoch: [192]  [120/312]  eta: 0:01:55  lr: 0.001290  min_lr: 0.001290  loss: 3.8790 (3.7543)  weight_decay: 0.0500 (0.0500)  time: 0.4897  data: 0.0917  max mem: 21002
Epoch: [192]  [130/312]  eta: 0:01:50  lr: 0.001289  min_lr: 0.001289  loss: 3.8670 (3.7602)  weight_decay: 0.0500 (0.0500)  time: 0.6732  data: 0.2126  max mem: 21002
Epoch: [192]  [140/312]  eta: 0:01:40  lr: 0.001289  min_lr: 0.001289  loss: 4.0390 (3.7818)  weight_decay: 0.0500 (0.0500)  time: 0.4718  data: 0.1267  max mem: 21002
Epoch: [192]  [150/312]  eta: 0:01:35  lr: 0.001288  min_lr: 0.001288  loss: 4.0907 (3.7950)  weight_decay: 0.0500 (0.0500)  time: 0.4871  data: 0.1318  max mem: 21002
Epoch: [192]  [160/312]  eta: 0:01:29  lr: 0.001287  min_lr: 0.001287  loss: 3.8245 (3.7894)  weight_decay: 0.0500 (0.0500)  time: 0.6159  data: 0.2234  max mem: 21002
Epoch: [192]  [170/312]  eta: 0:01:21  lr: 0.001286  min_lr: 0.001286  loss: 3.8013 (3.7925)  weight_decay: 0.0500 (0.0500)  time: 0.4442  data: 0.1212  max mem: 21002
Epoch: [192]  [180/312]  eta: 0:01:16  lr: 0.001286  min_lr: 0.001286  loss: 3.8364 (3.7895)  weight_decay: 0.0500 (0.0500)  time: 0.5165  data: 0.1766  max mem: 21002
Epoch: [192]  [190/312]  eta: 0:01:08  lr: 0.001285  min_lr: 0.001285  loss: 3.8167 (3.7838)  weight_decay: 0.0500 (0.0500)  time: 0.4862  data: 0.1476  max mem: 21002
Epoch: [192]  [200/312]  eta: 0:01:03  lr: 0.001284  min_lr: 0.001284  loss: 3.9653 (3.7927)  weight_decay: 0.0500 (0.0500)  time: 0.4545  data: 0.1464  max mem: 21002
Epoch: [192]  [210/312]  eta: 0:00:58  lr: 0.001284  min_lr: 0.001284  loss: 3.8082 (3.7824)  weight_decay: 0.0500 (0.0500)  time: 0.6675  data: 0.3588  max mem: 21002
Epoch: [192]  [220/312]  eta: 0:00:51  lr: 0.001283  min_lr: 0.001283  loss: 4.0406 (3.7894)  weight_decay: 0.0500 (0.0500)  time: 0.4975  data: 0.2130  max mem: 21002
Epoch: [192]  [230/312]  eta: 0:00:46  lr: 0.001282  min_lr: 0.001282  loss: 3.9705 (3.7888)  weight_decay: 0.0500 (0.0500)  time: 0.4793  data: 0.1940  max mem: 21002
Epoch: [192]  [240/312]  eta: 0:00:41  lr: 0.001282  min_lr: 0.001282  loss: 3.5772 (3.7824)  weight_decay: 0.0500 (0.0500)  time: 0.6792  data: 0.3916  max mem: 21002
Epoch: [192]  [250/312]  eta: 0:00:34  lr: 0.001281  min_lr: 0.001281  loss: 3.7431 (3.7862)  weight_decay: 0.0500 (0.0500)  time: 0.4847  data: 0.1994  max mem: 21002
Epoch: [192]  [260/312]  eta: 0:00:29  lr: 0.001280  min_lr: 0.001280  loss: 4.0431 (3.7910)  weight_decay: 0.0500 (0.0500)  time: 0.4788  data: 0.1942  max mem: 21002
Epoch: [192]  [270/312]  eta: 0:00:23  lr: 0.001280  min_lr: 0.001280  loss: 4.0550 (3.8027)  weight_decay: 0.0500 (0.0500)  time: 0.4797  data: 0.1929  max mem: 21002
Epoch: [192]  [280/312]  eta: 0:00:17  lr: 0.001279  min_lr: 0.001279  loss: 3.7773 (3.7900)  weight_decay: 0.0500 (0.0500)  time: 0.4583  data: 0.1716  max mem: 21002
Epoch: [192]  [290/312]  eta: 0:00:12  lr: 0.001278  min_lr: 0.001278  loss: 3.4477 (3.7876)  weight_decay: 0.0500 (0.0500)  time: 0.6132  data: 0.3288  max mem: 21002
Epoch: [192]  [300/312]  eta: 0:00:06  lr: 0.001278  min_lr: 0.001278  loss: 3.9121 (3.7892)  weight_decay: 0.0500 (0.0500)  time: 0.4381  data: 0.1577  max mem: 21002
Epoch: [192]  [310/312]  eta: 0:00:01  lr: 0.001277  min_lr: 0.001277  loss: 3.7944 (3.7935)  weight_decay: 0.0500 (0.0500)  time: 0.2816  data: 0.0050  max mem: 21002
Epoch: [192]  [311/312]  eta: 0:00:00  lr: 0.001277  min_lr: 0.001277  loss: 3.7944 (3.7923)  weight_decay: 0.0500 (0.0500)  time: 0.2816  data: 0.0050  max mem: 21002
Epoch: [192] Total time: 0:02:49 (0.5431 s / it)
Averaged stats: lr: 0.001277  min_lr: 0.001277  loss: 3.7944 (3.8010)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 1.4336 (1.4336)  acc1: 71.0938 (71.0938)  acc5: 90.1042 (90.1042)  time: 8.9478  data: 8.8291  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.7625 (1.6643)  acc1: 63.6719 (64.8000)  acc5: 86.5885 (86.7520)  time: 1.0795  data: 0.9811  max mem: 21002
Test: Total time: 0:00:09 (1.0926 s / it)
* Acc@1 65.116 Acc@5 86.672 loss 1.662
Accuracy of the model on the 50000 test images: 65.1%
Max accuracy: 65.12%
Epoch: [193]  [  0/312]  eta: 0:59:48  lr: 0.001277  min_lr: 0.001277  loss: 3.2973 (3.2973)  weight_decay: 0.0500 (0.0500)  time: 11.5005  data: 11.2005  max mem: 21002
Epoch: [193]  [ 10/312]  eta: 0:07:26  lr: 0.001276  min_lr: 0.001276  loss: 3.3256 (3.5984)  weight_decay: 0.0500 (0.0500)  time: 1.4798  data: 1.0460  max mem: 21002
Epoch: [193]  [ 20/312]  eta: 0:04:58  lr: 0.001276  min_lr: 0.001276  loss: 3.8071 (3.7713)  weight_decay: 0.0500 (0.0500)  time: 0.4973  data: 0.1296  max mem: 21002
Epoch: [193]  [ 30/312]  eta: 0:03:41  lr: 0.001275  min_lr: 0.001275  loss: 4.0342 (3.8389)  weight_decay: 0.0500 (0.0500)  time: 0.4035  data: 0.1146  max mem: 21002
Epoch: [193]  [ 40/312]  eta: 0:03:23  lr: 0.001274  min_lr: 0.001274  loss: 4.0194 (3.8412)  weight_decay: 0.0500 (0.0500)  time: 0.4605  data: 0.1690  max mem: 21002
Epoch: [193]  [ 50/312]  eta: 0:03:12  lr: 0.001274  min_lr: 0.001274  loss: 3.8554 (3.8199)  weight_decay: 0.0500 (0.0500)  time: 0.6567  data: 0.3667  max mem: 21002
Epoch: [193]  [ 60/312]  eta: 0:02:46  lr: 0.001273  min_lr: 0.001273  loss: 4.0419 (3.8605)  weight_decay: 0.0500 (0.0500)  time: 0.4843  data: 0.1984  max mem: 21002
Epoch: [193]  [ 70/312]  eta: 0:02:38  lr: 0.001272  min_lr: 0.001272  loss: 4.0693 (3.8976)  weight_decay: 0.0500 (0.0500)  time: 0.4558  data: 0.1592  max mem: 21002
Epoch: [193]  [ 80/312]  eta: 0:02:32  lr: 0.001272  min_lr: 0.001272  loss: 3.9503 (3.8939)  weight_decay: 0.0500 (0.0500)  time: 0.6520  data: 0.3551  max mem: 21002
Epoch: [193]  [ 90/312]  eta: 0:02:17  lr: 0.001271  min_lr: 0.001271  loss: 3.9387 (3.8952)  weight_decay: 0.0500 (0.0500)  time: 0.4825  data: 0.1965  max mem: 21002
Epoch: [193]  [100/312]  eta: 0:02:12  lr: 0.001270  min_lr: 0.001270  loss: 3.9003 (3.8734)  weight_decay: 0.0500 (0.0500)  time: 0.4900  data: 0.2043  max mem: 21002
Epoch: [193]  [110/312]  eta: 0:02:00  lr: 0.001270  min_lr: 0.001270  loss: 3.7904 (3.8782)  weight_decay: 0.0500 (0.0500)  time: 0.4900  data: 0.2043  max mem: 21002
Epoch: [193]  [120/312]  eta: 0:01:55  lr: 0.001269  min_lr: 0.001269  loss: 3.9167 (3.8650)  weight_decay: 0.0500 (0.0500)  time: 0.4832  data: 0.1973  max mem: 21002
Epoch: [193]  [130/312]  eta: 0:01:50  lr: 0.001268  min_lr: 0.001268  loss: 3.8963 (3.8676)  weight_decay: 0.0500 (0.0500)  time: 0.6799  data: 0.3936  max mem: 21002
Epoch: [193]  [140/312]  eta: 0:01:40  lr: 0.001268  min_lr: 0.001268  loss: 3.8655 (3.8668)  weight_decay: 0.0500 (0.0500)  time: 0.4831  data: 0.1969  max mem: 21002
Epoch: [193]  [150/312]  eta: 0:01:35  lr: 0.001267  min_lr: 0.001267  loss: 4.0210 (3.8898)  weight_decay: 0.0500 (0.0500)  time: 0.4810  data: 0.1946  max mem: 21002
Epoch: [193]  [160/312]  eta: 0:01:30  lr: 0.001266  min_lr: 0.001266  loss: 4.1889 (3.8778)  weight_decay: 0.0500 (0.0500)  time: 0.6606  data: 0.3743  max mem: 21002
Epoch: [193]  [170/312]  eta: 0:01:21  lr: 0.001266  min_lr: 0.001266  loss: 4.0633 (3.8766)  weight_decay: 0.0500 (0.0500)  time: 0.4642  data: 0.1803  max mem: 21002
Epoch: [193]  [180/312]  eta: 0:01:16  lr: 0.001265  min_lr: 0.001265  loss: 4.0633 (3.8776)  weight_decay: 0.0500 (0.0500)  time: 0.4787  data: 0.1940  max mem: 21002
Epoch: [193]  [190/312]  eta: 0:01:09  lr: 0.001264  min_lr: 0.001264  loss: 4.0728 (3.8810)  weight_decay: 0.0500 (0.0500)  time: 0.4793  data: 0.1940  max mem: 21002
Epoch: [193]  [200/312]  eta: 0:01:04  lr: 0.001264  min_lr: 0.001264  loss: 4.0791 (3.8853)  weight_decay: 0.0500 (0.0500)  time: 0.4883  data: 0.2017  max mem: 21002
Epoch: [193]  [210/312]  eta: 0:00:58  lr: 0.001263  min_lr: 0.001263  loss: 4.2136 (3.9028)  weight_decay: 0.0500 (0.0500)  time: 0.6936  data: 0.4029  max mem: 21002
Epoch: [193]  [220/312]  eta: 0:00:51  lr: 0.001262  min_lr: 0.001262  loss: 4.0113 (3.9005)  weight_decay: 0.0500 (0.0500)  time: 0.4918  data: 0.2019  max mem: 21002
Epoch: [193]  [230/312]  eta: 0:00:46  lr: 0.001262  min_lr: 0.001262  loss: 3.8386 (3.9041)  weight_decay: 0.0500 (0.0500)  time: 0.4889  data: 0.2014  max mem: 21002
Epoch: [193]  [240/312]  eta: 0:00:41  lr: 0.001261  min_lr: 0.001261  loss: 4.0428 (3.8982)  weight_decay: 0.0500 (0.0500)  time: 0.6880  data: 0.4007  max mem: 21002
Epoch: [193]  [250/312]  eta: 0:00:34  lr: 0.001260  min_lr: 0.001260  loss: 3.6252 (3.8868)  weight_decay: 0.0500 (0.0500)  time: 0.4862  data: 0.2000  max mem: 21002
Epoch: [193]  [260/312]  eta: 0:00:29  lr: 0.001260  min_lr: 0.001260  loss: 3.7720 (3.8797)  weight_decay: 0.0500 (0.0500)  time: 0.4673  data: 0.1815  max mem: 21002
Epoch: [193]  [270/312]  eta: 0:00:23  lr: 0.001259  min_lr: 0.001259  loss: 3.9519 (3.8812)  weight_decay: 0.0500 (0.0500)  time: 0.4692  data: 0.1815  max mem: 21002
Epoch: [193]  [280/312]  eta: 0:00:17  lr: 0.001258  min_lr: 0.001258  loss: 4.0274 (3.8820)  weight_decay: 0.0500 (0.0500)  time: 0.4764  data: 0.1899  max mem: 21002
Epoch: [193]  [290/312]  eta: 0:00:12  lr: 0.001258  min_lr: 0.001258  loss: 3.6068 (3.8691)  weight_decay: 0.0500 (0.0500)  time: 0.6127  data: 0.3274  max mem: 21002
Epoch: [193]  [300/312]  eta: 0:00:06  lr: 0.001257  min_lr: 0.001257  loss: 3.4837 (3.8609)  weight_decay: 0.0500 (0.0500)  time: 0.4197  data: 0.1378  max mem: 21002
Epoch: [193]  [310/312]  eta: 0:00:01  lr: 0.001256  min_lr: 0.001256  loss: 3.8924 (3.8610)  weight_decay: 0.0500 (0.0500)  time: 0.2774  data: 0.0001  max mem: 21002
Epoch: [193]  [311/312]  eta: 0:00:00  lr: 0.001256  min_lr: 0.001256  loss: 3.8924 (3.8613)  weight_decay: 0.0500 (0.0500)  time: 0.2773  data: 0.0001  max mem: 21002
Epoch: [193] Total time: 0:02:50 (0.5451 s / it)
Averaged stats: lr: 0.001256  min_lr: 0.001256  loss: 3.8924 (3.7965)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.5504 (1.5504)  acc1: 71.3542 (71.3542)  acc5: 91.4062 (91.4062)  time: 8.6917  data: 8.5735  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.9494 (1.8396)  acc1: 63.8021 (64.5600)  acc5: 85.9375 (86.4800)  time: 1.0509  data: 0.9527  max mem: 21002
Test: Total time: 0:00:09 (1.0613 s / it)
* Acc@1 64.366 Acc@5 86.394 loss 1.843
Accuracy of the model on the 50000 test images: 64.4%
Max accuracy: 65.12%
Epoch: [194]  [  0/312]  eta: 1:03:19  lr: 0.001256  min_lr: 0.001256  loss: 3.7694 (3.7694)  weight_decay: 0.0500 (0.0500)  time: 12.1780  data: 10.9782  max mem: 21002
Epoch: [194]  [ 10/312]  eta: 0:08:30  lr: 0.001255  min_lr: 0.001255  loss: 4.1025 (3.8242)  weight_decay: 0.0500 (0.0500)  time: 1.6905  data: 1.0120  max mem: 21002
Epoch: [194]  [ 20/312]  eta: 0:05:10  lr: 0.001255  min_lr: 0.001255  loss: 3.9985 (3.8168)  weight_decay: 0.0500 (0.0500)  time: 0.5062  data: 0.0122  max mem: 21002
Epoch: [194]  [ 30/312]  eta: 0:03:49  lr: 0.001254  min_lr: 0.001254  loss: 3.7735 (3.7881)  weight_decay: 0.0500 (0.0500)  time: 0.3302  data: 0.0050  max mem: 21002
Epoch: [194]  [ 40/312]  eta: 0:03:07  lr: 0.001253  min_lr: 0.001253  loss: 3.7229 (3.7429)  weight_decay: 0.0500 (0.0500)  time: 0.2999  data: 0.0011  max mem: 21002
Epoch: [194]  [ 50/312]  eta: 0:02:58  lr: 0.001253  min_lr: 0.001253  loss: 3.3267 (3.6615)  weight_decay: 0.0500 (0.0500)  time: 0.4813  data: 0.1111  max mem: 21002
Epoch: [194]  [ 60/312]  eta: 0:02:38  lr: 0.001252  min_lr: 0.001252  loss: 3.3594 (3.6457)  weight_decay: 0.0500 (0.0500)  time: 0.5047  data: 0.1425  max mem: 21002
Epoch: [194]  [ 70/312]  eta: 0:02:31  lr: 0.001251  min_lr: 0.001251  loss: 3.5588 (3.6451)  weight_decay: 0.0500 (0.0500)  time: 0.4833  data: 0.1370  max mem: 21002
Epoch: [194]  [ 80/312]  eta: 0:02:25  lr: 0.001251  min_lr: 0.001251  loss: 3.5710 (3.6321)  weight_decay: 0.0500 (0.0500)  time: 0.6220  data: 0.2369  max mem: 21002
Epoch: [194]  [ 90/312]  eta: 0:02:11  lr: 0.001250  min_lr: 0.001250  loss: 3.9135 (3.6809)  weight_decay: 0.0500 (0.0500)  time: 0.4762  data: 0.1322  max mem: 21002
Epoch: [194]  [100/312]  eta: 0:02:07  lr: 0.001249  min_lr: 0.001249  loss: 3.9703 (3.6697)  weight_decay: 0.0500 (0.0500)  time: 0.5033  data: 0.1279  max mem: 21002
Epoch: [194]  [110/312]  eta: 0:01:56  lr: 0.001249  min_lr: 0.001249  loss: 3.8182 (3.6908)  weight_decay: 0.0500 (0.0500)  time: 0.4877  data: 0.1279  max mem: 21002
Epoch: [194]  [120/312]  eta: 0:01:51  lr: 0.001248  min_lr: 0.001248  loss: 3.9460 (3.7062)  weight_decay: 0.0500 (0.0500)  time: 0.4642  data: 0.1159  max mem: 21002
Epoch: [194]  [130/312]  eta: 0:01:47  lr: 0.001247  min_lr: 0.001247  loss: 4.0404 (3.7355)  weight_decay: 0.0500 (0.0500)  time: 0.6663  data: 0.2362  max mem: 21002
Epoch: [194]  [140/312]  eta: 0:01:37  lr: 0.001247  min_lr: 0.001247  loss: 3.9221 (3.7247)  weight_decay: 0.0500 (0.0500)  time: 0.5040  data: 0.1210  max mem: 21002
Epoch: [194]  [150/312]  eta: 0:01:33  lr: 0.001246  min_lr: 0.001246  loss: 3.7126 (3.7272)  weight_decay: 0.0500 (0.0500)  time: 0.5000  data: 0.0972  max mem: 21002
Epoch: [194]  [160/312]  eta: 0:01:27  lr: 0.001245  min_lr: 0.001245  loss: 3.8106 (3.7320)  weight_decay: 0.0500 (0.0500)  time: 0.6252  data: 0.1867  max mem: 21002
Epoch: [194]  [170/312]  eta: 0:01:20  lr: 0.001245  min_lr: 0.001245  loss: 3.9555 (3.7453)  weight_decay: 0.0500 (0.0500)  time: 0.4749  data: 0.0903  max mem: 21002
Epoch: [194]  [180/312]  eta: 0:01:15  lr: 0.001244  min_lr: 0.001244  loss: 3.9300 (3.7359)  weight_decay: 0.0500 (0.0500)  time: 0.5309  data: 0.0975  max mem: 21002
Epoch: [194]  [190/312]  eta: 0:01:07  lr: 0.001243  min_lr: 0.001243  loss: 3.8922 (3.7429)  weight_decay: 0.0500 (0.0500)  time: 0.4848  data: 0.0976  max mem: 21002
Epoch: [194]  [200/312]  eta: 0:01:02  lr: 0.001243  min_lr: 0.001243  loss: 3.8922 (3.7429)  weight_decay: 0.0500 (0.0500)  time: 0.4778  data: 0.1055  max mem: 21002
Epoch: [194]  [210/312]  eta: 0:00:57  lr: 0.001242  min_lr: 0.001242  loss: 3.8491 (3.7553)  weight_decay: 0.0500 (0.0500)  time: 0.6765  data: 0.2335  max mem: 21002
Epoch: [194]  [220/312]  eta: 0:00:51  lr: 0.001241  min_lr: 0.001241  loss: 3.8173 (3.7484)  weight_decay: 0.0500 (0.0500)  time: 0.4869  data: 0.1291  max mem: 21002
Epoch: [194]  [230/312]  eta: 0:00:45  lr: 0.001241  min_lr: 0.001241  loss: 4.0237 (3.7624)  weight_decay: 0.0500 (0.0500)  time: 0.4812  data: 0.1392  max mem: 21002
Epoch: [194]  [240/312]  eta: 0:00:40  lr: 0.001240  min_lr: 0.001240  loss: 4.0569 (3.7583)  weight_decay: 0.0500 (0.0500)  time: 0.6259  data: 0.2769  max mem: 21002
Epoch: [194]  [250/312]  eta: 0:00:34  lr: 0.001239  min_lr: 0.001239  loss: 3.9282 (3.7610)  weight_decay: 0.0500 (0.0500)  time: 0.4901  data: 0.1388  max mem: 21002
Epoch: [194]  [260/312]  eta: 0:00:29  lr: 0.001239  min_lr: 0.001239  loss: 3.8069 (3.7547)  weight_decay: 0.0500 (0.0500)  time: 0.5973  data: 0.1319  max mem: 21002
Epoch: [194]  [270/312]  eta: 0:00:23  lr: 0.001238  min_lr: 0.001238  loss: 3.7739 (3.7537)  weight_decay: 0.0500 (0.0500)  time: 0.5402  data: 0.1318  max mem: 21002
Epoch: [194]  [280/312]  eta: 0:00:17  lr: 0.001237  min_lr: 0.001237  loss: 3.7413 (3.7454)  weight_decay: 0.0500 (0.0500)  time: 0.4728  data: 0.0692  max mem: 21002
Epoch: [194]  [290/312]  eta: 0:00:12  lr: 0.001237  min_lr: 0.001237  loss: 3.8781 (3.7534)  weight_decay: 0.0500 (0.0500)  time: 0.6286  data: 0.1429  max mem: 21002
Epoch: [194]  [300/312]  eta: 0:00:06  lr: 0.001236  min_lr: 0.001236  loss: 4.0101 (3.7496)  weight_decay: 0.0500 (0.0500)  time: 0.4400  data: 0.0741  max mem: 21002
Epoch: [194]  [310/312]  eta: 0:00:01  lr: 0.001235  min_lr: 0.001235  loss: 3.7303 (3.7542)  weight_decay: 0.0500 (0.0500)  time: 0.2775  data: 0.0001  max mem: 21002
Epoch: [194]  [311/312]  eta: 0:00:00  lr: 0.001235  min_lr: 0.001235  loss: 3.7303 (3.7554)  weight_decay: 0.0500 (0.0500)  time: 0.2774  data: 0.0001  max mem: 21002
Epoch: [194] Total time: 0:02:49 (0.5435 s / it)
Averaged stats: lr: 0.001235  min_lr: 0.001235  loss: 3.7303 (3.8011)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:23  loss: 1.4876 (1.4876)  acc1: 72.7865 (72.7865)  acc5: 90.7552 (90.7552)  time: 9.2467  data: 9.1281  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.9474 (1.8391)  acc1: 63.9323 (65.1840)  acc5: 86.4583 (86.3680)  time: 1.1130  data: 1.0143  max mem: 21002
Test: Total time: 0:00:10 (1.1270 s / it)
* Acc@1 64.890 Acc@5 86.890 loss 1.830
Accuracy of the model on the 50000 test images: 64.9%
Max accuracy: 65.12%
Epoch: [195]  [  0/312]  eta: 0:54:07  lr: 0.001235  min_lr: 0.001235  loss: 3.8809 (3.8809)  weight_decay: 0.0500 (0.0500)  time: 10.4095  data: 9.3041  max mem: 21002
Epoch: [195]  [ 10/312]  eta: 0:07:23  lr: 0.001235  min_lr: 0.001235  loss: 3.7823 (3.7627)  weight_decay: 0.0500 (0.0500)  time: 1.4689  data: 1.0904  max mem: 21002
Epoch: [195]  [ 20/312]  eta: 0:05:17  lr: 0.001234  min_lr: 0.001234  loss: 3.6219 (3.5999)  weight_decay: 0.0500 (0.0500)  time: 0.6202  data: 0.2332  max mem: 21002
Epoch: [195]  [ 30/312]  eta: 0:03:53  lr: 0.001233  min_lr: 0.001233  loss: 3.6219 (3.6464)  weight_decay: 0.0500 (0.0500)  time: 0.4783  data: 0.0990  max mem: 21002
Epoch: [195]  [ 40/312]  eta: 0:03:39  lr: 0.001233  min_lr: 0.001233  loss: 3.9045 (3.6647)  weight_decay: 0.0500 (0.0500)  time: 0.5147  data: 0.1004  max mem: 21002
Epoch: [195]  [ 50/312]  eta: 0:03:27  lr: 0.001232  min_lr: 0.001232  loss: 3.9314 (3.7028)  weight_decay: 0.0500 (0.0500)  time: 0.7324  data: 0.1984  max mem: 21002
Epoch: [195]  [ 60/312]  eta: 0:02:58  lr: 0.001231  min_lr: 0.001231  loss: 3.9314 (3.7244)  weight_decay: 0.0500 (0.0500)  time: 0.5077  data: 0.0988  max mem: 21002
Epoch: [195]  [ 70/312]  eta: 0:02:43  lr: 0.001231  min_lr: 0.001231  loss: 3.7779 (3.6997)  weight_decay: 0.0500 (0.0500)  time: 0.3796  data: 0.0007  max mem: 21002
Epoch: [195]  [ 80/312]  eta: 0:02:29  lr: 0.001230  min_lr: 0.001230  loss: 3.9179 (3.7283)  weight_decay: 0.0500 (0.0500)  time: 0.4422  data: 0.0008  max mem: 21002
Epoch: [195]  [ 90/312]  eta: 0:02:20  lr: 0.001229  min_lr: 0.001229  loss: 4.0858 (3.7401)  weight_decay: 0.0500 (0.0500)  time: 0.4916  data: 0.0590  max mem: 21002
Epoch: [195]  [100/312]  eta: 0:02:18  lr: 0.001229  min_lr: 0.001229  loss: 4.0333 (3.7605)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.1225  max mem: 21002
Epoch: [195]  [110/312]  eta: 0:02:05  lr: 0.001228  min_lr: 0.001228  loss: 3.9890 (3.7744)  weight_decay: 0.0500 (0.0500)  time: 0.5464  data: 0.0648  max mem: 21002
Epoch: [195]  [120/312]  eta: 0:01:59  lr: 0.001227  min_lr: 0.001227  loss: 4.1793 (3.8174)  weight_decay: 0.0500 (0.0500)  time: 0.4862  data: 0.0322  max mem: 21002
Epoch: [195]  [130/312]  eta: 0:01:54  lr: 0.001227  min_lr: 0.001227  loss: 4.1793 (3.8214)  weight_decay: 0.0500 (0.0500)  time: 0.6998  data: 0.0461  max mem: 21002
Epoch: [195]  [140/312]  eta: 0:01:44  lr: 0.001226  min_lr: 0.001226  loss: 3.6900 (3.8240)  weight_decay: 0.0500 (0.0500)  time: 0.5060  data: 0.0149  max mem: 21002
Epoch: [195]  [150/312]  eta: 0:01:39  lr: 0.001225  min_lr: 0.001225  loss: 3.8238 (3.8156)  weight_decay: 0.0500 (0.0500)  time: 0.4782  data: 0.0542  max mem: 21002
Epoch: [195]  [160/312]  eta: 0:01:29  lr: 0.001225  min_lr: 0.001225  loss: 3.5737 (3.8089)  weight_decay: 0.0500 (0.0500)  time: 0.4815  data: 0.0541  max mem: 21002
Epoch: [195]  [170/312]  eta: 0:01:24  lr: 0.001224  min_lr: 0.001224  loss: 3.8740 (3.8135)  weight_decay: 0.0500 (0.0500)  time: 0.4856  data: 0.0342  max mem: 21002
Epoch: [195]  [180/312]  eta: 0:01:20  lr: 0.001223  min_lr: 0.001223  loss: 3.8740 (3.8129)  weight_decay: 0.0500 (0.0500)  time: 0.7369  data: 0.0395  max mem: 21002
Epoch: [195]  [190/312]  eta: 0:01:12  lr: 0.001223  min_lr: 0.001223  loss: 4.1172 (3.8302)  weight_decay: 0.0500 (0.0500)  time: 0.5439  data: 0.0059  max mem: 21002
Epoch: [195]  [200/312]  eta: 0:01:05  lr: 0.001222  min_lr: 0.001222  loss: 4.1522 (3.8370)  weight_decay: 0.0500 (0.0500)  time: 0.4122  data: 0.0200  max mem: 21002
Epoch: [195]  [210/312]  eta: 0:01:00  lr: 0.001221  min_lr: 0.001221  loss: 3.9931 (3.8190)  weight_decay: 0.0500 (0.0500)  time: 0.6027  data: 0.0201  max mem: 21002
Epoch: [195]  [220/312]  eta: 0:00:53  lr: 0.001221  min_lr: 0.001221  loss: 3.4733 (3.8042)  weight_decay: 0.0500 (0.0500)  time: 0.4793  data: 0.0007  max mem: 21002
Epoch: [195]  [230/312]  eta: 0:00:47  lr: 0.001220  min_lr: 0.001220  loss: 3.8192 (3.8072)  weight_decay: 0.0500 (0.0500)  time: 0.4874  data: 0.0009  max mem: 21002
Epoch: [195]  [240/312]  eta: 0:00:41  lr: 0.001219  min_lr: 0.001219  loss: 3.9203 (3.8064)  weight_decay: 0.0500 (0.0500)  time: 0.4911  data: 0.0010  max mem: 21002
Epoch: [195]  [250/312]  eta: 0:00:35  lr: 0.001219  min_lr: 0.001219  loss: 3.8645 (3.8058)  weight_decay: 0.0500 (0.0500)  time: 0.5311  data: 0.0094  max mem: 21002
Epoch: [195]  [260/312]  eta: 0:00:30  lr: 0.001218  min_lr: 0.001218  loss: 3.9867 (3.8103)  weight_decay: 0.0500 (0.0500)  time: 0.7679  data: 0.0179  max mem: 21002
Epoch: [195]  [270/312]  eta: 0:00:24  lr: 0.001217  min_lr: 0.001217  loss: 4.0727 (3.8155)  weight_decay: 0.0500 (0.0500)  time: 0.5285  data: 0.0092  max mem: 21002
Epoch: [195]  [280/312]  eta: 0:00:18  lr: 0.001217  min_lr: 0.001217  loss: 3.6840 (3.8060)  weight_decay: 0.0500 (0.0500)  time: 0.4144  data: 0.0093  max mem: 21002
Epoch: [195]  [290/312]  eta: 0:00:12  lr: 0.001216  min_lr: 0.001216  loss: 3.8917 (3.8121)  weight_decay: 0.0500 (0.0500)  time: 0.5284  data: 0.0091  max mem: 21002
Epoch: [195]  [300/312]  eta: 0:00:06  lr: 0.001215  min_lr: 0.001215  loss: 4.1307 (3.8102)  weight_decay: 0.0500 (0.0500)  time: 0.3948  data: 0.0002  max mem: 21002
Epoch: [195]  [310/312]  eta: 0:00:01  lr: 0.001215  min_lr: 0.001215  loss: 3.8568 (3.8054)  weight_decay: 0.0500 (0.0500)  time: 0.2769  data: 0.0001  max mem: 21002
Epoch: [195]  [311/312]  eta: 0:00:00  lr: 0.001215  min_lr: 0.001215  loss: 3.8568 (3.8047)  weight_decay: 0.0500 (0.0500)  time: 0.2768  data: 0.0001  max mem: 21002
Epoch: [195] Total time: 0:02:53 (0.5557 s / it)
Averaged stats: lr: 0.001215  min_lr: 0.001215  loss: 3.8568 (3.7983)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:21  loss: 1.4826 (1.4826)  acc1: 72.3958 (72.3958)  acc5: 91.1458 (91.1458)  time: 9.0695  data: 8.9519  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.9016 (1.7686)  acc1: 62.2396 (63.5840)  acc5: 85.4167 (86.2720)  time: 1.0965  data: 0.9947  max mem: 21002
Test: Total time: 0:00:10 (1.1395 s / it)
* Acc@1 63.670 Acc@5 86.188 loss 1.768
Accuracy of the model on the 50000 test images: 63.7%
Max accuracy: 65.12%
Epoch: [196]  [  0/312]  eta: 0:58:29  lr: 0.001215  min_lr: 0.001215  loss: 4.1501 (4.1501)  weight_decay: 0.0500 (0.0500)  time: 11.2498  data: 8.8349  max mem: 21002
Epoch: [196]  [ 10/312]  eta: 0:07:51  lr: 0.001214  min_lr: 0.001214  loss: 3.4934 (3.7382)  weight_decay: 0.0500 (0.0500)  time: 1.5606  data: 1.0296  max mem: 21002
Epoch: [196]  [ 20/312]  eta: 0:05:31  lr: 0.001213  min_lr: 0.001213  loss: 3.6328 (3.7424)  weight_decay: 0.0500 (0.0500)  time: 0.6282  data: 0.1822  max mem: 21002
Epoch: [196]  [ 30/312]  eta: 0:04:03  lr: 0.001213  min_lr: 0.001213  loss: 4.0224 (3.8078)  weight_decay: 0.0500 (0.0500)  time: 0.4810  data: 0.0581  max mem: 21002
Epoch: [196]  [ 40/312]  eta: 0:03:30  lr: 0.001212  min_lr: 0.001212  loss: 3.6152 (3.7244)  weight_decay: 0.0500 (0.0500)  time: 0.3938  data: 0.0008  max mem: 21002
Epoch: [196]  [ 50/312]  eta: 0:03:20  lr: 0.001211  min_lr: 0.001211  loss: 3.6538 (3.7276)  weight_decay: 0.0500 (0.0500)  time: 0.6100  data: 0.0360  max mem: 21002
Epoch: [196]  [ 60/312]  eta: 0:02:54  lr: 0.001211  min_lr: 0.001211  loss: 3.7025 (3.7579)  weight_decay: 0.0500 (0.0500)  time: 0.5244  data: 0.0502  max mem: 21002
Epoch: [196]  [ 70/312]  eta: 0:02:46  lr: 0.001210  min_lr: 0.001210  loss: 3.8564 (3.7520)  weight_decay: 0.0500 (0.0500)  time: 0.4886  data: 0.0617  max mem: 21002
Epoch: [196]  [ 80/312]  eta: 0:02:37  lr: 0.001209  min_lr: 0.001209  loss: 3.8679 (3.7694)  weight_decay: 0.0500 (0.0500)  time: 0.6363  data: 0.0483  max mem: 21002
Epoch: [196]  [ 90/312]  eta: 0:02:21  lr: 0.001209  min_lr: 0.001209  loss: 3.7587 (3.7528)  weight_decay: 0.0500 (0.0500)  time: 0.4642  data: 0.0163  max mem: 21002
Epoch: [196]  [100/312]  eta: 0:02:17  lr: 0.001208  min_lr: 0.001208  loss: 3.7628 (3.7615)  weight_decay: 0.0500 (0.0500)  time: 0.5185  data: 0.0661  max mem: 21002
Epoch: [196]  [110/312]  eta: 0:02:04  lr: 0.001207  min_lr: 0.001207  loss: 3.6681 (3.7444)  weight_decay: 0.0500 (0.0500)  time: 0.5079  data: 0.0514  max mem: 21002
Epoch: [196]  [120/312]  eta: 0:01:59  lr: 0.001207  min_lr: 0.001207  loss: 3.6072 (3.7469)  weight_decay: 0.0500 (0.0500)  time: 0.4916  data: 0.0378  max mem: 21002
Epoch: [196]  [130/312]  eta: 0:01:52  lr: 0.001206  min_lr: 0.001206  loss: 4.0878 (3.7720)  weight_decay: 0.0500 (0.0500)  time: 0.6499  data: 0.0539  max mem: 21002
Epoch: [196]  [140/312]  eta: 0:01:42  lr: 0.001205  min_lr: 0.001205  loss: 4.1070 (3.7786)  weight_decay: 0.0500 (0.0500)  time: 0.4632  data: 0.0323  max mem: 21002
Epoch: [196]  [150/312]  eta: 0:01:37  lr: 0.001205  min_lr: 0.001205  loss: 3.9826 (3.7911)  weight_decay: 0.0500 (0.0500)  time: 0.4998  data: 0.0491  max mem: 21002
Epoch: [196]  [160/312]  eta: 0:01:32  lr: 0.001204  min_lr: 0.001204  loss: 3.8819 (3.7954)  weight_decay: 0.0500 (0.0500)  time: 0.6781  data: 0.0529  max mem: 21002
Epoch: [196]  [170/312]  eta: 0:01:24  lr: 0.001203  min_lr: 0.001203  loss: 3.8267 (3.7850)  weight_decay: 0.0500 (0.0500)  time: 0.5083  data: 0.0491  max mem: 21002
Epoch: [196]  [180/312]  eta: 0:01:18  lr: 0.001203  min_lr: 0.001203  loss: 3.9672 (3.8017)  weight_decay: 0.0500 (0.0500)  time: 0.5036  data: 0.0340  max mem: 21002
Epoch: [196]  [190/312]  eta: 0:01:10  lr: 0.001202  min_lr: 0.001202  loss: 3.9672 (3.8038)  weight_decay: 0.0500 (0.0500)  time: 0.4763  data: 0.0051  max mem: 21002
Epoch: [196]  [200/312]  eta: 0:01:06  lr: 0.001201  min_lr: 0.001201  loss: 3.9278 (3.8157)  weight_decay: 0.0500 (0.0500)  time: 0.5411  data: 0.0483  max mem: 21002
Epoch: [196]  [210/312]  eta: 0:01:00  lr: 0.001201  min_lr: 0.001201  loss: 3.9672 (3.8150)  weight_decay: 0.0500 (0.0500)  time: 0.6842  data: 0.0483  max mem: 21002
Epoch: [196]  [220/312]  eta: 0:00:53  lr: 0.001200  min_lr: 0.001200  loss: 3.9590 (3.8236)  weight_decay: 0.0500 (0.0500)  time: 0.4321  data: 0.0007  max mem: 21002
Epoch: [196]  [230/312]  eta: 0:00:47  lr: 0.001199  min_lr: 0.001199  loss: 4.0654 (3.8205)  weight_decay: 0.0500 (0.0500)  time: 0.4700  data: 0.0007  max mem: 21002
Epoch: [196]  [240/312]  eta: 0:00:42  lr: 0.001199  min_lr: 0.001199  loss: 4.0213 (3.8233)  weight_decay: 0.0500 (0.0500)  time: 0.6934  data: 0.0007  max mem: 21002
Epoch: [196]  [250/312]  eta: 0:00:35  lr: 0.001198  min_lr: 0.001198  loss: 3.9056 (3.8262)  weight_decay: 0.0500 (0.0500)  time: 0.5433  data: 0.0303  max mem: 21002
Epoch: [196]  [260/312]  eta: 0:00:30  lr: 0.001197  min_lr: 0.001197  loss: 3.8548 (3.8242)  weight_decay: 0.0500 (0.0500)  time: 0.5252  data: 0.0595  max mem: 21002
Epoch: [196]  [270/312]  eta: 0:00:23  lr: 0.001197  min_lr: 0.001197  loss: 3.8748 (3.8195)  weight_decay: 0.0500 (0.0500)  time: 0.4998  data: 0.0346  max mem: 21002
Epoch: [196]  [280/312]  eta: 0:00:18  lr: 0.001196  min_lr: 0.001196  loss: 3.8727 (3.8180)  weight_decay: 0.0500 (0.0500)  time: 0.4911  data: 0.0261  max mem: 21002
Epoch: [196]  [290/312]  eta: 0:00:12  lr: 0.001196  min_lr: 0.001196  loss: 3.9290 (3.8253)  weight_decay: 0.0500 (0.0500)  time: 0.5750  data: 0.0382  max mem: 21002
Epoch: [196]  [300/312]  eta: 0:00:06  lr: 0.001195  min_lr: 0.001195  loss: 4.1144 (3.8293)  weight_decay: 0.0500 (0.0500)  time: 0.3720  data: 0.0173  max mem: 21002
Epoch: [196]  [310/312]  eta: 0:00:01  lr: 0.001194  min_lr: 0.001194  loss: 3.9929 (3.8353)  weight_decay: 0.0500 (0.0500)  time: 0.2770  data: 0.0001  max mem: 21002
Epoch: [196]  [311/312]  eta: 0:00:00  lr: 0.001194  min_lr: 0.001194  loss: 3.9929 (3.8358)  weight_decay: 0.0500 (0.0500)  time: 0.2770  data: 0.0001  max mem: 21002
Epoch: [196] Total time: 0:02:53 (0.5554 s / it)
Averaged stats: lr: 0.001194  min_lr: 0.001194  loss: 3.9929 (3.7915)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.6824 (1.6824)  acc1: 73.1771 (73.1771)  acc5: 90.7552 (90.7552)  time: 8.8083  data: 8.6895  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 2.1923 (2.0166)  acc1: 58.7240 (62.0320)  acc5: 83.0729 (84.6400)  time: 1.0652  data: 0.9656  max mem: 21002
Test: Total time: 0:00:09 (1.0965 s / it)
* Acc@1 61.992 Acc@5 84.776 loss 2.027
Accuracy of the model on the 50000 test images: 62.0%
Max accuracy: 65.12%
Epoch: [197]  [  0/312]  eta: 1:06:21  lr: 0.001194  min_lr: 0.001194  loss: 3.8870 (3.8870)  weight_decay: 0.0500 (0.0500)  time: 12.7598  data: 8.7190  max mem: 21002
Epoch: [197]  [ 10/312]  eta: 0:08:41  lr: 0.001193  min_lr: 0.001193  loss: 3.7538 (3.6370)  weight_decay: 0.0500 (0.0500)  time: 1.7276  data: 1.1039  max mem: 21002
Epoch: [197]  [ 20/312]  eta: 0:05:12  lr: 0.001193  min_lr: 0.001193  loss: 3.9575 (3.7721)  weight_decay: 0.0500 (0.0500)  time: 0.4868  data: 0.1837  max mem: 21002
Epoch: [197]  [ 30/312]  eta: 0:03:51  lr: 0.001192  min_lr: 0.001192  loss: 4.0268 (3.7911)  weight_decay: 0.0500 (0.0500)  time: 0.3225  data: 0.0128  max mem: 21002
Epoch: [197]  [ 40/312]  eta: 0:03:09  lr: 0.001191  min_lr: 0.001191  loss: 3.9581 (3.7988)  weight_decay: 0.0500 (0.0500)  time: 0.3025  data: 0.0006  max mem: 21002
Epoch: [197]  [ 50/312]  eta: 0:03:00  lr: 0.001191  min_lr: 0.001191  loss: 3.9581 (3.8103)  weight_decay: 0.0500 (0.0500)  time: 0.4843  data: 0.0760  max mem: 21002
Epoch: [197]  [ 60/312]  eta: 0:02:41  lr: 0.001190  min_lr: 0.001190  loss: 3.7582 (3.7981)  weight_decay: 0.0500 (0.0500)  time: 0.5248  data: 0.0823  max mem: 21002
Epoch: [197]  [ 70/312]  eta: 0:02:36  lr: 0.001189  min_lr: 0.001189  loss: 3.7116 (3.7547)  weight_decay: 0.0500 (0.0500)  time: 0.5403  data: 0.0779  max mem: 21002
Epoch: [197]  [ 80/312]  eta: 0:02:25  lr: 0.001189  min_lr: 0.001189  loss: 3.8462 (3.7847)  weight_decay: 0.0500 (0.0500)  time: 0.5837  data: 0.0885  max mem: 21002
Epoch: [197]  [ 90/312]  eta: 0:02:16  lr: 0.001188  min_lr: 0.001188  loss: 4.0908 (3.7910)  weight_decay: 0.0500 (0.0500)  time: 0.4913  data: 0.0684  max mem: 21002
Epoch: [197]  [100/312]  eta: 0:02:12  lr: 0.001187  min_lr: 0.001187  loss: 3.9548 (3.7732)  weight_decay: 0.0500 (0.0500)  time: 0.6109  data: 0.1104  max mem: 21002
Epoch: [197]  [110/312]  eta: 0:02:00  lr: 0.001187  min_lr: 0.001187  loss: 3.9728 (3.7842)  weight_decay: 0.0500 (0.0500)  time: 0.5200  data: 0.0736  max mem: 21002
Epoch: [197]  [120/312]  eta: 0:01:55  lr: 0.001186  min_lr: 0.001186  loss: 3.9231 (3.7697)  weight_decay: 0.0500 (0.0500)  time: 0.5001  data: 0.0668  max mem: 21002
Epoch: [197]  [130/312]  eta: 0:01:50  lr: 0.001186  min_lr: 0.001186  loss: 3.8722 (3.7734)  weight_decay: 0.0500 (0.0500)  time: 0.6546  data: 0.1020  max mem: 21002
Epoch: [197]  [140/312]  eta: 0:01:41  lr: 0.001185  min_lr: 0.001185  loss: 3.8722 (3.7626)  weight_decay: 0.0500 (0.0500)  time: 0.5145  data: 0.0738  max mem: 21002
Epoch: [197]  [150/312]  eta: 0:01:35  lr: 0.001184  min_lr: 0.001184  loss: 3.6943 (3.7604)  weight_decay: 0.0500 (0.0500)  time: 0.5077  data: 0.0681  max mem: 21002
Epoch: [197]  [160/312]  eta: 0:01:28  lr: 0.001184  min_lr: 0.001184  loss: 3.7026 (3.7496)  weight_decay: 0.0500 (0.0500)  time: 0.5167  data: 0.0746  max mem: 21002
Epoch: [197]  [170/312]  eta: 0:01:22  lr: 0.001183  min_lr: 0.001183  loss: 3.7106 (3.7603)  weight_decay: 0.0500 (0.0500)  time: 0.5036  data: 0.1206  max mem: 21002
Epoch: [197]  [180/312]  eta: 0:01:17  lr: 0.001182  min_lr: 0.001182  loss: 3.8050 (3.7630)  weight_decay: 0.0500 (0.0500)  time: 0.6296  data: 0.1417  max mem: 21002
Epoch: [197]  [190/312]  eta: 0:01:09  lr: 0.001182  min_lr: 0.001182  loss: 3.9541 (3.7626)  weight_decay: 0.0500 (0.0500)  time: 0.5024  data: 0.0715  max mem: 21002
Epoch: [197]  [200/312]  eta: 0:01:04  lr: 0.001181  min_lr: 0.001181  loss: 3.8127 (3.7557)  weight_decay: 0.0500 (0.0500)  time: 0.5000  data: 0.0860  max mem: 21002
Epoch: [197]  [210/312]  eta: 0:00:59  lr: 0.001180  min_lr: 0.001180  loss: 3.8127 (3.7523)  weight_decay: 0.0500 (0.0500)  time: 0.6441  data: 0.1177  max mem: 21002
Epoch: [197]  [220/312]  eta: 0:00:52  lr: 0.001180  min_lr: 0.001180  loss: 3.9055 (3.7605)  weight_decay: 0.0500 (0.0500)  time: 0.5241  data: 0.0965  max mem: 21002
Epoch: [197]  [230/312]  eta: 0:00:47  lr: 0.001179  min_lr: 0.001179  loss: 3.9239 (3.7680)  weight_decay: 0.0500 (0.0500)  time: 0.5290  data: 0.1201  max mem: 21002
Epoch: [197]  [240/312]  eta: 0:00:40  lr: 0.001178  min_lr: 0.001178  loss: 3.8627 (3.7713)  weight_decay: 0.0500 (0.0500)  time: 0.4749  data: 0.0869  max mem: 21002
Epoch: [197]  [250/312]  eta: 0:00:35  lr: 0.001178  min_lr: 0.001178  loss: 3.8627 (3.7754)  weight_decay: 0.0500 (0.0500)  time: 0.4952  data: 0.0933  max mem: 21002
Epoch: [197]  [260/312]  eta: 0:00:29  lr: 0.001177  min_lr: 0.001177  loss: 3.9558 (3.7735)  weight_decay: 0.0500 (0.0500)  time: 0.6961  data: 0.1429  max mem: 21002
Epoch: [197]  [270/312]  eta: 0:00:23  lr: 0.001176  min_lr: 0.001176  loss: 3.6576 (3.7656)  weight_decay: 0.0500 (0.0500)  time: 0.5445  data: 0.0674  max mem: 21002
Epoch: [197]  [280/312]  eta: 0:00:18  lr: 0.001176  min_lr: 0.001176  loss: 3.8487 (3.7747)  weight_decay: 0.0500 (0.0500)  time: 0.5389  data: 0.0756  max mem: 21002
Epoch: [197]  [290/312]  eta: 0:00:12  lr: 0.001175  min_lr: 0.001175  loss: 3.7427 (3.7687)  weight_decay: 0.0500 (0.0500)  time: 0.5784  data: 0.1145  max mem: 21002
Epoch: [197]  [300/312]  eta: 0:00:06  lr: 0.001174  min_lr: 0.001174  loss: 3.8934 (3.7811)  weight_decay: 0.0500 (0.0500)  time: 0.3773  data: 0.0557  max mem: 21002
Epoch: [197]  [310/312]  eta: 0:00:01  lr: 0.001174  min_lr: 0.001174  loss: 4.0873 (3.7909)  weight_decay: 0.0500 (0.0500)  time: 0.2956  data: 0.0111  max mem: 21002
Epoch: [197]  [311/312]  eta: 0:00:00  lr: 0.001174  min_lr: 0.001174  loss: 4.0873 (3.7935)  weight_decay: 0.0500 (0.0500)  time: 0.2844  data: 0.0001  max mem: 21002
Epoch: [197] Total time: 0:02:52 (0.5537 s / it)
Averaged stats: lr: 0.001174  min_lr: 0.001174  loss: 4.0873 (3.7989)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:23  loss: 1.6352 (1.6352)  acc1: 69.1406 (69.1406)  acc5: 89.1927 (89.1927)  time: 9.2608  data: 9.1429  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 2.0134 (1.9224)  acc1: 61.8490 (61.3920)  acc5: 83.7240 (84.4480)  time: 1.1185  data: 1.0159  max mem: 21002
Test: Total time: 0:00:10 (1.1390 s / it)
* Acc@1 60.992 Acc@5 84.430 loss 1.937
Accuracy of the model on the 50000 test images: 61.0%
Max accuracy: 65.12%
Epoch: [198]  [  0/312]  eta: 1:03:05  lr: 0.001174  min_lr: 0.001174  loss: 3.9752 (3.9752)  weight_decay: 0.0500 (0.0500)  time: 12.1317  data: 10.2066  max mem: 21002
Epoch: [198]  [ 10/312]  eta: 0:08:12  lr: 0.001173  min_lr: 0.001173  loss: 3.5113 (3.5129)  weight_decay: 0.0500 (0.0500)  time: 1.6296  data: 1.1246  max mem: 21002
Epoch: [198]  [ 20/312]  eta: 0:05:34  lr: 0.001172  min_lr: 0.001172  loss: 3.5113 (3.5810)  weight_decay: 0.0500 (0.0500)  time: 0.5954  data: 0.2132  max mem: 21002
Epoch: [198]  [ 30/312]  eta: 0:04:05  lr: 0.001172  min_lr: 0.001172  loss: 3.5028 (3.5049)  weight_decay: 0.0500 (0.0500)  time: 0.4515  data: 0.1055  max mem: 21002
Epoch: [198]  [ 40/312]  eta: 0:03:36  lr: 0.001171  min_lr: 0.001171  loss: 3.6734 (3.5815)  weight_decay: 0.0500 (0.0500)  time: 0.4311  data: 0.0976  max mem: 21002
Epoch: [198]  [ 50/312]  eta: 0:03:21  lr: 0.001170  min_lr: 0.001170  loss: 3.7814 (3.5531)  weight_decay: 0.0500 (0.0500)  time: 0.6100  data: 0.2111  max mem: 21002
Epoch: [198]  [ 60/312]  eta: 0:02:53  lr: 0.001170  min_lr: 0.001170  loss: 3.7405 (3.6092)  weight_decay: 0.0500 (0.0500)  time: 0.4686  data: 0.1144  max mem: 21002
Epoch: [198]  [ 70/312]  eta: 0:02:47  lr: 0.001169  min_lr: 0.001169  loss: 3.7827 (3.6400)  weight_decay: 0.0500 (0.0500)  time: 0.5046  data: 0.1169  max mem: 21002
Epoch: [198]  [ 80/312]  eta: 0:02:35  lr: 0.001168  min_lr: 0.001168  loss: 3.7239 (3.6357)  weight_decay: 0.0500 (0.0500)  time: 0.6073  data: 0.1228  max mem: 21002
Epoch: [198]  [ 90/312]  eta: 0:02:25  lr: 0.001168  min_lr: 0.001168  loss: 3.7239 (3.6411)  weight_decay: 0.0500 (0.0500)  time: 0.5283  data: 0.1442  max mem: 21002
Epoch: [198]  [100/312]  eta: 0:02:19  lr: 0.001167  min_lr: 0.001167  loss: 3.7652 (3.6797)  weight_decay: 0.0500 (0.0500)  time: 0.6221  data: 0.2429  max mem: 21002
Epoch: [198]  [110/312]  eta: 0:02:06  lr: 0.001166  min_lr: 0.001166  loss: 3.9628 (3.6924)  weight_decay: 0.0500 (0.0500)  time: 0.4875  data: 0.1054  max mem: 21002
Epoch: [198]  [120/312]  eta: 0:02:01  lr: 0.001166  min_lr: 0.001166  loss: 3.9628 (3.7172)  weight_decay: 0.0500 (0.0500)  time: 0.4892  data: 0.1219  max mem: 21002
Epoch: [198]  [130/312]  eta: 0:01:55  lr: 0.001165  min_lr: 0.001165  loss: 3.9767 (3.7165)  weight_decay: 0.0500 (0.0500)  time: 0.6927  data: 0.2442  max mem: 21002
Epoch: [198]  [140/312]  eta: 0:01:45  lr: 0.001164  min_lr: 0.001164  loss: 3.7218 (3.7214)  weight_decay: 0.0500 (0.0500)  time: 0.4957  data: 0.1230  max mem: 21002
Epoch: [198]  [150/312]  eta: 0:01:40  lr: 0.001164  min_lr: 0.001164  loss: 3.8626 (3.7246)  weight_decay: 0.0500 (0.0500)  time: 0.4898  data: 0.1356  max mem: 21002
Epoch: [198]  [160/312]  eta: 0:01:32  lr: 0.001163  min_lr: 0.001163  loss: 3.9515 (3.7367)  weight_decay: 0.0500 (0.0500)  time: 0.5613  data: 0.1609  max mem: 21002
Epoch: [198]  [170/312]  eta: 0:01:25  lr: 0.001162  min_lr: 0.001162  loss: 3.9765 (3.7385)  weight_decay: 0.0500 (0.0500)  time: 0.4656  data: 0.1312  max mem: 21002
Epoch: [198]  [180/312]  eta: 0:01:19  lr: 0.001162  min_lr: 0.001162  loss: 3.6891 (3.7254)  weight_decay: 0.0500 (0.0500)  time: 0.5925  data: 0.2232  max mem: 21002
Epoch: [198]  [190/312]  eta: 0:01:11  lr: 0.001161  min_lr: 0.001161  loss: 3.6891 (3.7335)  weight_decay: 0.0500 (0.0500)  time: 0.4912  data: 0.1180  max mem: 21002
Epoch: [198]  [200/312]  eta: 0:01:06  lr: 0.001161  min_lr: 0.001161  loss: 4.1217 (3.7468)  weight_decay: 0.0500 (0.0500)  time: 0.4650  data: 0.1252  max mem: 21002
Epoch: [198]  [210/312]  eta: 0:01:00  lr: 0.001160  min_lr: 0.001160  loss: 3.9197 (3.7381)  weight_decay: 0.0500 (0.0500)  time: 0.6700  data: 0.2894  max mem: 21002
Epoch: [198]  [220/312]  eta: 0:00:53  lr: 0.001159  min_lr: 0.001159  loss: 3.5707 (3.7349)  weight_decay: 0.0500 (0.0500)  time: 0.4999  data: 0.1654  max mem: 21002
Epoch: [198]  [230/312]  eta: 0:00:48  lr: 0.001159  min_lr: 0.001159  loss: 3.8341 (3.7395)  weight_decay: 0.0500 (0.0500)  time: 0.5019  data: 0.1434  max mem: 21002
Epoch: [198]  [240/312]  eta: 0:00:41  lr: 0.001158  min_lr: 0.001158  loss: 4.0026 (3.7457)  weight_decay: 0.0500 (0.0500)  time: 0.5614  data: 0.1489  max mem: 21002
Epoch: [198]  [250/312]  eta: 0:00:36  lr: 0.001157  min_lr: 0.001157  loss: 3.6138 (3.7323)  weight_decay: 0.0500 (0.0500)  time: 0.5246  data: 0.1782  max mem: 21002
Epoch: [198]  [260/312]  eta: 0:00:30  lr: 0.001157  min_lr: 0.001157  loss: 3.8197 (3.7404)  weight_decay: 0.0500 (0.0500)  time: 0.6631  data: 0.3312  max mem: 21002
Epoch: [198]  [270/312]  eta: 0:00:24  lr: 0.001156  min_lr: 0.001156  loss: 4.0256 (3.7442)  weight_decay: 0.0500 (0.0500)  time: 0.4906  data: 0.1598  max mem: 21002
Epoch: [198]  [280/312]  eta: 0:00:18  lr: 0.001155  min_lr: 0.001155  loss: 3.9235 (3.7396)  weight_decay: 0.0500 (0.0500)  time: 0.4986  data: 0.1695  max mem: 21002
Epoch: [198]  [290/312]  eta: 0:00:12  lr: 0.001155  min_lr: 0.001155  loss: 3.5161 (3.7347)  weight_decay: 0.0500 (0.0500)  time: 0.5987  data: 0.2520  max mem: 21002
Epoch: [198]  [300/312]  eta: 0:00:06  lr: 0.001154  min_lr: 0.001154  loss: 3.9366 (3.7376)  weight_decay: 0.0500 (0.0500)  time: 0.3844  data: 0.0830  max mem: 21002
Epoch: [198]  [310/312]  eta: 0:00:01  lr: 0.001153  min_lr: 0.001153  loss: 3.9366 (3.7362)  weight_decay: 0.0500 (0.0500)  time: 0.2770  data: 0.0001  max mem: 21002
Epoch: [198]  [311/312]  eta: 0:00:00  lr: 0.001153  min_lr: 0.001153  loss: 3.9366 (3.7362)  weight_decay: 0.0500 (0.0500)  time: 0.2769  data: 0.0001  max mem: 21002
Epoch: [198] Total time: 0:02:55 (0.5612 s / it)
Averaged stats: lr: 0.001153  min_lr: 0.001153  loss: 3.9366 (3.7996)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.5282 (1.5282)  acc1: 71.4844 (71.4844)  acc5: 90.3646 (90.3646)  time: 8.7553  data: 8.6373  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.9388 (1.7852)  acc1: 60.6771 (63.3440)  acc5: 84.7656 (85.8080)  time: 1.0891  data: 0.9913  max mem: 21002
Test: Total time: 0:00:09 (1.0991 s / it)
* Acc@1 63.380 Acc@5 85.766 loss 1.796
Accuracy of the model on the 50000 test images: 63.4%
Max accuracy: 65.12%
Epoch: [199]  [  0/312]  eta: 1:06:14  lr: 0.001153  min_lr: 0.001153  loss: 3.9370 (3.9370)  weight_decay: 0.0500 (0.0500)  time: 12.7389  data: 11.0115  max mem: 21002
Epoch: [199]  [ 10/312]  eta: 0:07:57  lr: 0.001153  min_lr: 0.001153  loss: 3.9370 (3.7451)  weight_decay: 0.0500 (0.0500)  time: 1.5823  data: 1.0775  max mem: 21002
Epoch: [199]  [ 20/312]  eta: 0:05:53  lr: 0.001152  min_lr: 0.001152  loss: 3.8905 (3.7569)  weight_decay: 0.0500 (0.0500)  time: 0.6325  data: 0.1418  max mem: 21002
Epoch: [199]  [ 30/312]  eta: 0:04:16  lr: 0.001151  min_lr: 0.001151  loss: 3.9854 (3.8179)  weight_decay: 0.0500 (0.0500)  time: 0.5402  data: 0.1000  max mem: 21002
Epoch: [199]  [ 40/312]  eta: 0:03:29  lr: 0.001151  min_lr: 0.001151  loss: 3.9854 (3.8417)  weight_decay: 0.0500 (0.0500)  time: 0.3059  data: 0.0039  max mem: 21002
Epoch: [199]  [ 50/312]  eta: 0:03:10  lr: 0.001150  min_lr: 0.001150  loss: 3.7799 (3.7851)  weight_decay: 0.0500 (0.0500)  time: 0.4463  data: 0.0042  max mem: 21002
Epoch: [199]  [ 60/312]  eta: 0:02:46  lr: 0.001149  min_lr: 0.001149  loss: 3.7799 (3.7691)  weight_decay: 0.0500 (0.0500)  time: 0.4422  data: 0.0187  max mem: 21002
Epoch: [199]  [ 70/312]  eta: 0:02:44  lr: 0.001149  min_lr: 0.001149  loss: 4.0534 (3.7912)  weight_decay: 0.0500 (0.0500)  time: 0.5528  data: 0.0640  max mem: 21002
Epoch: [199]  [ 80/312]  eta: 0:02:30  lr: 0.001148  min_lr: 0.001148  loss: 4.0814 (3.8045)  weight_decay: 0.0500 (0.0500)  time: 0.6073  data: 0.0463  max mem: 21002
Epoch: [199]  [ 90/312]  eta: 0:02:18  lr: 0.001147  min_lr: 0.001147  loss: 4.0204 (3.7759)  weight_decay: 0.0500 (0.0500)  time: 0.4384  data: 0.0057  max mem: 21002
Epoch: [199]  [100/312]  eta: 0:02:16  lr: 0.001147  min_lr: 0.001147  loss: 4.0204 (3.8082)  weight_decay: 0.0500 (0.0500)  time: 0.6199  data: 0.0339  max mem: 21002
Epoch: [199]  [110/312]  eta: 0:02:03  lr: 0.001146  min_lr: 0.001146  loss: 4.1146 (3.8176)  weight_decay: 0.0500 (0.0500)  time: 0.5410  data: 0.0289  max mem: 21002
Epoch: [199]  [120/312]  eta: 0:01:57  lr: 0.001145  min_lr: 0.001145  loss: 4.0210 (3.8125)  weight_decay: 0.0500 (0.0500)  time: 0.4698  data: 0.0403  max mem: 21002
Epoch: [199]  [130/312]  eta: 0:01:52  lr: 0.001145  min_lr: 0.001145  loss: 3.9811 (3.8126)  weight_decay: 0.0500 (0.0500)  time: 0.6622  data: 0.0404  max mem: 21002
Epoch: [199]  [140/312]  eta: 0:01:43  lr: 0.001144  min_lr: 0.001144  loss: 4.0683 (3.8150)  weight_decay: 0.0500 (0.0500)  time: 0.5399  data: 0.0639  max mem: 21002
Epoch: [199]  [150/312]  eta: 0:01:38  lr: 0.001143  min_lr: 0.001143  loss: 4.0763 (3.8107)  weight_decay: 0.0500 (0.0500)  time: 0.5175  data: 0.1239  max mem: 21002
Epoch: [199]  [160/312]  eta: 0:01:30  lr: 0.001143  min_lr: 0.001143  loss: 4.0589 (3.8144)  weight_decay: 0.0500 (0.0500)  time: 0.5348  data: 0.0607  max mem: 21002
Epoch: [199]  [170/312]  eta: 0:01:23  lr: 0.001142  min_lr: 0.001142  loss: 3.8263 (3.7981)  weight_decay: 0.0500 (0.0500)  time: 0.4640  data: 0.0214  max mem: 21002
Epoch: [199]  [180/312]  eta: 0:01:18  lr: 0.001142  min_lr: 0.001142  loss: 3.8263 (3.8037)  weight_decay: 0.0500 (0.0500)  time: 0.5933  data: 0.0381  max mem: 21002
Epoch: [199]  [190/312]  eta: 0:01:10  lr: 0.001141  min_lr: 0.001141  loss: 4.0518 (3.8123)  weight_decay: 0.0500 (0.0500)  time: 0.5202  data: 0.0175  max mem: 21002
Epoch: [199]  [200/312]  eta: 0:01:05  lr: 0.001140  min_lr: 0.001140  loss: 3.9881 (3.8148)  weight_decay: 0.0500 (0.0500)  time: 0.5196  data: 0.0519  max mem: 21002
Epoch: [199]  [210/312]  eta: 0:01:00  lr: 0.001140  min_lr: 0.001140  loss: 4.0026 (3.8174)  weight_decay: 0.0500 (0.0500)  time: 0.6607  data: 0.0521  max mem: 21002
Epoch: [199]  [220/312]  eta: 0:00:53  lr: 0.001139  min_lr: 0.001139  loss: 4.0026 (3.8234)  weight_decay: 0.0500 (0.0500)  time: 0.5049  data: 0.0069  max mem: 21002
Epoch: [199]  [230/312]  eta: 0:00:48  lr: 0.001138  min_lr: 0.001138  loss: 3.8974 (3.8268)  weight_decay: 0.0500 (0.0500)  time: 0.5541  data: 0.0642  max mem: 21002
Epoch: [199]  [240/312]  eta: 0:00:41  lr: 0.001138  min_lr: 0.001138  loss: 3.8238 (3.8155)  weight_decay: 0.0500 (0.0500)  time: 0.5863  data: 0.0582  max mem: 21002
Epoch: [199]  [250/312]  eta: 0:00:35  lr: 0.001137  min_lr: 0.001137  loss: 3.7372 (3.8235)  weight_decay: 0.0500 (0.0500)  time: 0.4849  data: 0.0437  max mem: 21002
Epoch: [199]  [260/312]  eta: 0:00:30  lr: 0.001136  min_lr: 0.001136  loss: 3.9547 (3.8218)  weight_decay: 0.0500 (0.0500)  time: 0.5748  data: 0.0863  max mem: 21002
Epoch: [199]  [270/312]  eta: 0:00:24  lr: 0.001136  min_lr: 0.001136  loss: 3.9555 (3.8267)  weight_decay: 0.0500 (0.0500)  time: 0.4956  data: 0.0434  max mem: 21002
Epoch: [199]  [280/312]  eta: 0:00:18  lr: 0.001135  min_lr: 0.001135  loss: 4.0119 (3.8253)  weight_decay: 0.0500 (0.0500)  time: 0.5332  data: 0.0695  max mem: 21002
Epoch: [199]  [290/312]  eta: 0:00:12  lr: 0.001134  min_lr: 0.001134  loss: 3.9426 (3.8335)  weight_decay: 0.0500 (0.0500)  time: 0.5764  data: 0.0692  max mem: 21002
Epoch: [199]  [300/312]  eta: 0:00:06  lr: 0.001134  min_lr: 0.001134  loss: 3.9974 (3.8361)  weight_decay: 0.0500 (0.0500)  time: 0.3755  data: 0.0035  max mem: 21002
Epoch: [199]  [310/312]  eta: 0:00:01  lr: 0.001133  min_lr: 0.001133  loss: 4.0204 (3.8358)  weight_decay: 0.0500 (0.0500)  time: 0.2908  data: 0.0034  max mem: 21002
Epoch: [199]  [311/312]  eta: 0:00:00  lr: 0.001133  min_lr: 0.001133  loss: 4.0204 (3.8386)  weight_decay: 0.0500 (0.0500)  time: 0.2880  data: 0.0001  max mem: 21002
Epoch: [199] Total time: 0:02:53 (0.5568 s / it)
Averaged stats: lr: 0.001133  min_lr: 0.001133  loss: 4.0204 (3.8145)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.5540 (1.5540)  acc1: 69.1406 (69.1406)  acc5: 90.3646 (90.3646)  time: 8.8336  data: 8.7148  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.8866 (1.7930)  acc1: 63.5417 (64.0960)  acc5: 85.8073 (85.6160)  time: 1.2225  data: 1.1260  max mem: 21002
Test: Total time: 0:00:11 (1.2331 s / it)
* Acc@1 63.924 Acc@5 86.090 loss 1.798
Accuracy of the model on the 50000 test images: 63.9%
Max accuracy: 65.12%
Epoch: [200]  [  0/312]  eta: 1:05:30  lr: 0.001133  min_lr: 0.001133  loss: 3.8477 (3.8477)  weight_decay: 0.0500 (0.0500)  time: 12.5973  data: 10.8452  max mem: 21002
Epoch: [200]  [ 10/312]  eta: 0:09:02  lr: 0.001132  min_lr: 0.001132  loss: 4.0023 (3.9353)  weight_decay: 0.0500 (0.0500)  time: 1.7963  data: 1.0675  max mem: 21002
Epoch: [200]  [ 20/312]  eta: 0:05:29  lr: 0.001132  min_lr: 0.001132  loss: 4.0878 (4.0155)  weight_decay: 0.0500 (0.0500)  time: 0.5541  data: 0.0497  max mem: 21002
Epoch: [200]  [ 30/312]  eta: 0:04:04  lr: 0.001131  min_lr: 0.001131  loss: 3.9155 (3.9220)  weight_decay: 0.0500 (0.0500)  time: 0.3571  data: 0.0235  max mem: 21002
Epoch: [200]  [ 40/312]  eta: 0:03:17  lr: 0.001130  min_lr: 0.001130  loss: 3.9155 (3.9061)  weight_decay: 0.0500 (0.0500)  time: 0.3048  data: 0.0191  max mem: 21002
Epoch: [200]  [ 50/312]  eta: 0:02:56  lr: 0.001130  min_lr: 0.001130  loss: 3.8912 (3.8517)  weight_decay: 0.0500 (0.0500)  time: 0.3692  data: 0.0008  max mem: 21002
Epoch: [200]  [ 60/312]  eta: 0:02:41  lr: 0.001129  min_lr: 0.001129  loss: 3.9204 (3.8725)  weight_decay: 0.0500 (0.0500)  time: 0.4708  data: 0.0777  max mem: 21002
Epoch: [200]  [ 70/312]  eta: 0:02:33  lr: 0.001128  min_lr: 0.001128  loss: 4.0447 (3.8543)  weight_decay: 0.0500 (0.0500)  time: 0.5428  data: 0.1074  max mem: 21002
Epoch: [200]  [ 80/312]  eta: 0:02:22  lr: 0.001128  min_lr: 0.001128  loss: 3.9007 (3.8164)  weight_decay: 0.0500 (0.0500)  time: 0.5323  data: 0.1204  max mem: 21002
Epoch: [200]  [ 90/312]  eta: 0:02:14  lr: 0.001127  min_lr: 0.001127  loss: 3.7108 (3.8052)  weight_decay: 0.0500 (0.0500)  time: 0.5041  data: 0.0907  max mem: 21002
Epoch: [200]  [100/312]  eta: 0:02:10  lr: 0.001126  min_lr: 0.001126  loss: 3.7939 (3.8028)  weight_decay: 0.0500 (0.0500)  time: 0.6070  data: 0.0757  max mem: 21002
Epoch: [200]  [110/312]  eta: 0:02:01  lr: 0.001126  min_lr: 0.001126  loss: 3.8995 (3.8034)  weight_decay: 0.0500 (0.0500)  time: 0.5701  data: 0.1577  max mem: 21002
Epoch: [200]  [120/312]  eta: 0:01:53  lr: 0.001125  min_lr: 0.001125  loss: 3.9605 (3.7978)  weight_decay: 0.0500 (0.0500)  time: 0.4722  data: 0.0954  max mem: 21002
Epoch: [200]  [130/312]  eta: 0:01:48  lr: 0.001125  min_lr: 0.001125  loss: 3.9110 (3.7973)  weight_decay: 0.0500 (0.0500)  time: 0.5912  data: 0.1208  max mem: 21002
Epoch: [200]  [140/312]  eta: 0:01:42  lr: 0.001124  min_lr: 0.001124  loss: 3.9223 (3.7979)  weight_decay: 0.0500 (0.0500)  time: 0.6262  data: 0.2297  max mem: 21002
Epoch: [200]  [150/312]  eta: 0:01:34  lr: 0.001123  min_lr: 0.001123  loss: 3.8688 (3.7859)  weight_decay: 0.0500 (0.0500)  time: 0.4998  data: 0.1222  max mem: 21002
Epoch: [200]  [160/312]  eta: 0:01:29  lr: 0.001123  min_lr: 0.001123  loss: 3.8374 (3.7903)  weight_decay: 0.0500 (0.0500)  time: 0.5236  data: 0.1372  max mem: 21002
Epoch: [200]  [170/312]  eta: 0:01:21  lr: 0.001122  min_lr: 0.001122  loss: 3.9011 (3.7856)  weight_decay: 0.0500 (0.0500)  time: 0.5033  data: 0.1381  max mem: 21002
Epoch: [200]  [180/312]  eta: 0:01:16  lr: 0.001121  min_lr: 0.001121  loss: 3.8438 (3.7748)  weight_decay: 0.0500 (0.0500)  time: 0.5463  data: 0.1449  max mem: 21002
Epoch: [200]  [190/312]  eta: 0:01:11  lr: 0.001121  min_lr: 0.001121  loss: 3.5704 (3.7576)  weight_decay: 0.0500 (0.0500)  time: 0.6614  data: 0.2930  max mem: 21002
Epoch: [200]  [200/312]  eta: 0:01:04  lr: 0.001120  min_lr: 0.001120  loss: 3.6168 (3.7571)  weight_decay: 0.0500 (0.0500)  time: 0.5036  data: 0.1503  max mem: 21002
Epoch: [200]  [210/312]  eta: 0:00:58  lr: 0.001119  min_lr: 0.001119  loss: 3.7346 (3.7551)  weight_decay: 0.0500 (0.0500)  time: 0.5098  data: 0.1740  max mem: 21002
Epoch: [200]  [220/312]  eta: 0:00:53  lr: 0.001119  min_lr: 0.001119  loss: 3.6790 (3.7467)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.3770  max mem: 21002
Epoch: [200]  [230/312]  eta: 0:00:46  lr: 0.001118  min_lr: 0.001118  loss: 3.7196 (3.7467)  weight_decay: 0.0500 (0.0500)  time: 0.4985  data: 0.2044  max mem: 21002
Epoch: [200]  [240/312]  eta: 0:00:41  lr: 0.001117  min_lr: 0.001117  loss: 3.8623 (3.7532)  weight_decay: 0.0500 (0.0500)  time: 0.5114  data: 0.2170  max mem: 21002
Epoch: [200]  [250/312]  eta: 0:00:35  lr: 0.001117  min_lr: 0.001117  loss: 3.8926 (3.7511)  weight_decay: 0.0500 (0.0500)  time: 0.5066  data: 0.2172  max mem: 21002
Epoch: [200]  [260/312]  eta: 0:00:29  lr: 0.001116  min_lr: 0.001116  loss: 3.9524 (3.7640)  weight_decay: 0.0500 (0.0500)  time: 0.4843  data: 0.1963  max mem: 21002
Epoch: [200]  [270/312]  eta: 0:00:24  lr: 0.001115  min_lr: 0.001115  loss: 4.1254 (3.7663)  weight_decay: 0.0500 (0.0500)  time: 0.7136  data: 0.4261  max mem: 21002
Epoch: [200]  [280/312]  eta: 0:00:18  lr: 0.001115  min_lr: 0.001115  loss: 3.9853 (3.7703)  weight_decay: 0.0500 (0.0500)  time: 0.5205  data: 0.2325  max mem: 21002
Epoch: [200]  [290/312]  eta: 0:00:12  lr: 0.001114  min_lr: 0.001114  loss: 3.6386 (3.7629)  weight_decay: 0.0500 (0.0500)  time: 0.5045  data: 0.2178  max mem: 21002
Epoch: [200]  [300/312]  eta: 0:00:06  lr: 0.001114  min_lr: 0.001114  loss: 3.9554 (3.7752)  weight_decay: 0.0500 (0.0500)  time: 0.5293  data: 0.2484  max mem: 21002
Epoch: [200]  [310/312]  eta: 0:00:01  lr: 0.001113  min_lr: 0.001113  loss: 4.0613 (3.7774)  weight_decay: 0.0500 (0.0500)  time: 0.3097  data: 0.0328  max mem: 21002
Epoch: [200]  [311/312]  eta: 0:00:00  lr: 0.001113  min_lr: 0.001113  loss: 4.0346 (3.7777)  weight_decay: 0.0500 (0.0500)  time: 0.3096  data: 0.0328  max mem: 21002
Epoch: [200] Total time: 0:02:54 (0.5578 s / it)
Averaged stats: lr: 0.001113  min_lr: 0.001113  loss: 4.0346 (3.7765)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:27  loss: 1.3601 (1.3601)  acc1: 75.5208 (75.5208)  acc5: 92.8385 (92.8385)  time: 9.7563  data: 9.6375  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.8377 (1.7516)  acc1: 63.0208 (65.2640)  acc5: 86.5885 (86.9280)  time: 1.1693  data: 1.0709  max mem: 21002
Test: Total time: 0:00:10 (1.1804 s / it)
* Acc@1 65.396 Acc@5 86.940 loss 1.761
Accuracy of the model on the 50000 test images: 65.4%
Max accuracy: 65.40%
Epoch: [201]  [  0/312]  eta: 1:05:10  lr: 0.001113  min_lr: 0.001113  loss: 3.6972 (3.6972)  weight_decay: 0.0500 (0.0500)  time: 12.5327  data: 11.1358  max mem: 21002
Epoch: [201]  [ 10/312]  eta: 0:08:26  lr: 0.001112  min_lr: 0.001112  loss: 3.8623 (3.8790)  weight_decay: 0.0500 (0.0500)  time: 1.6760  data: 1.0134  max mem: 21002
Epoch: [201]  [ 20/312]  eta: 0:05:05  lr: 0.001112  min_lr: 0.001112  loss: 3.9351 (3.8665)  weight_decay: 0.0500 (0.0500)  time: 0.4719  data: 0.0359  max mem: 21002
Epoch: [201]  [ 30/312]  eta: 0:03:46  lr: 0.001111  min_lr: 0.001111  loss: 3.8633 (3.8094)  weight_decay: 0.0500 (0.0500)  time: 0.3224  data: 0.0357  max mem: 21002
Epoch: [201]  [ 40/312]  eta: 0:03:08  lr: 0.001110  min_lr: 0.001110  loss: 3.6941 (3.7734)  weight_decay: 0.0500 (0.0500)  time: 0.3254  data: 0.0008  max mem: 21002
Epoch: [201]  [ 50/312]  eta: 0:02:59  lr: 0.001110  min_lr: 0.001110  loss: 3.8396 (3.7766)  weight_decay: 0.0500 (0.0500)  time: 0.4980  data: 0.1414  max mem: 21002
Epoch: [201]  [ 60/312]  eta: 0:02:37  lr: 0.001109  min_lr: 0.001109  loss: 3.8386 (3.7620)  weight_decay: 0.0500 (0.0500)  time: 0.4852  data: 0.1651  max mem: 21002
Epoch: [201]  [ 70/312]  eta: 0:02:35  lr: 0.001108  min_lr: 0.001108  loss: 4.0086 (3.7906)  weight_decay: 0.0500 (0.0500)  time: 0.5331  data: 0.1659  max mem: 21002
Epoch: [201]  [ 80/312]  eta: 0:02:24  lr: 0.001108  min_lr: 0.001108  loss: 3.9117 (3.7663)  weight_decay: 0.0500 (0.0500)  time: 0.6105  data: 0.2471  max mem: 21002
Epoch: [201]  [ 90/312]  eta: 0:02:16  lr: 0.001107  min_lr: 0.001107  loss: 3.9117 (3.7749)  weight_decay: 0.0500 (0.0500)  time: 0.5165  data: 0.1204  max mem: 21002
Epoch: [201]  [100/312]  eta: 0:02:09  lr: 0.001106  min_lr: 0.001106  loss: 3.8502 (3.7527)  weight_decay: 0.0500 (0.0500)  time: 0.5598  data: 0.0908  max mem: 21002
Epoch: [201]  [110/312]  eta: 0:02:00  lr: 0.001106  min_lr: 0.001106  loss: 3.6280 (3.7491)  weight_decay: 0.0500 (0.0500)  time: 0.5275  data: 0.1164  max mem: 21002
Epoch: [201]  [120/312]  eta: 0:01:53  lr: 0.001105  min_lr: 0.001105  loss: 3.7150 (3.7557)  weight_decay: 0.0500 (0.0500)  time: 0.4952  data: 0.1135  max mem: 21002
Epoch: [201]  [130/312]  eta: 0:01:48  lr: 0.001104  min_lr: 0.001104  loss: 3.6991 (3.7477)  weight_decay: 0.0500 (0.0500)  time: 0.5682  data: 0.1728  max mem: 21002
Epoch: [201]  [140/312]  eta: 0:01:40  lr: 0.001104  min_lr: 0.001104  loss: 3.7016 (3.7647)  weight_decay: 0.0500 (0.0500)  time: 0.5509  data: 0.1341  max mem: 21002
Epoch: [201]  [150/312]  eta: 0:01:34  lr: 0.001103  min_lr: 0.001103  loss: 3.8887 (3.7753)  weight_decay: 0.0500 (0.0500)  time: 0.5072  data: 0.1155  max mem: 21002
Epoch: [201]  [160/312]  eta: 0:01:28  lr: 0.001102  min_lr: 0.001102  loss: 3.8184 (3.7686)  weight_decay: 0.0500 (0.0500)  time: 0.5434  data: 0.1642  max mem: 21002
Epoch: [201]  [170/312]  eta: 0:01:21  lr: 0.001102  min_lr: 0.001102  loss: 3.8295 (3.7828)  weight_decay: 0.0500 (0.0500)  time: 0.5308  data: 0.1387  max mem: 21002
Epoch: [201]  [180/312]  eta: 0:01:16  lr: 0.001101  min_lr: 0.001101  loss: 3.8295 (3.7772)  weight_decay: 0.0500 (0.0500)  time: 0.5598  data: 0.1300  max mem: 21002
Epoch: [201]  [190/312]  eta: 0:01:09  lr: 0.001101  min_lr: 0.001101  loss: 3.8182 (3.7889)  weight_decay: 0.0500 (0.0500)  time: 0.5430  data: 0.1448  max mem: 21002
Epoch: [201]  [200/312]  eta: 0:01:04  lr: 0.001100  min_lr: 0.001100  loss: 3.8082 (3.7770)  weight_decay: 0.0500 (0.0500)  time: 0.5248  data: 0.1345  max mem: 21002
Epoch: [201]  [210/312]  eta: 0:00:58  lr: 0.001099  min_lr: 0.001099  loss: 3.8082 (3.7868)  weight_decay: 0.0500 (0.0500)  time: 0.5685  data: 0.1418  max mem: 21002
Epoch: [201]  [220/312]  eta: 0:00:52  lr: 0.001099  min_lr: 0.001099  loss: 4.0145 (3.7877)  weight_decay: 0.0500 (0.0500)  time: 0.5088  data: 0.1495  max mem: 21002
Epoch: [201]  [230/312]  eta: 0:00:46  lr: 0.001098  min_lr: 0.001098  loss: 3.8725 (3.7988)  weight_decay: 0.0500 (0.0500)  time: 0.5361  data: 0.1681  max mem: 21002
Epoch: [201]  [240/312]  eta: 0:00:40  lr: 0.001097  min_lr: 0.001097  loss: 3.8725 (3.7934)  weight_decay: 0.0500 (0.0500)  time: 0.5451  data: 0.1841  max mem: 21002
Epoch: [201]  [250/312]  eta: 0:00:34  lr: 0.001097  min_lr: 0.001097  loss: 3.9714 (3.8021)  weight_decay: 0.0500 (0.0500)  time: 0.5047  data: 0.1412  max mem: 21002
Epoch: [201]  [260/312]  eta: 0:00:29  lr: 0.001096  min_lr: 0.001096  loss: 4.1127 (3.8034)  weight_decay: 0.0500 (0.0500)  time: 0.5790  data: 0.1343  max mem: 21002
Epoch: [201]  [270/312]  eta: 0:00:23  lr: 0.001095  min_lr: 0.001095  loss: 3.8601 (3.8075)  weight_decay: 0.0500 (0.0500)  time: 0.5180  data: 0.1459  max mem: 21002
Epoch: [201]  [280/312]  eta: 0:00:18  lr: 0.001095  min_lr: 0.001095  loss: 3.8601 (3.8075)  weight_decay: 0.0500 (0.0500)  time: 0.5285  data: 0.1266  max mem: 21002
Epoch: [201]  [290/312]  eta: 0:00:12  lr: 0.001094  min_lr: 0.001094  loss: 3.8808 (3.8063)  weight_decay: 0.0500 (0.0500)  time: 0.6016  data: 0.0945  max mem: 21002
Epoch: [201]  [300/312]  eta: 0:00:06  lr: 0.001094  min_lr: 0.001094  loss: 3.9220 (3.8115)  weight_decay: 0.0500 (0.0500)  time: 0.4600  data: 0.0735  max mem: 21002
Epoch: [201]  [310/312]  eta: 0:00:01  lr: 0.001093  min_lr: 0.001093  loss: 4.1031 (3.8181)  weight_decay: 0.0500 (0.0500)  time: 0.3160  data: 0.0404  max mem: 21002
Epoch: [201]  [311/312]  eta: 0:00:00  lr: 0.001093  min_lr: 0.001093  loss: 4.0226 (3.8183)  weight_decay: 0.0500 (0.0500)  time: 0.3126  data: 0.0364  max mem: 21002
Epoch: [201] Total time: 0:02:51 (0.5504 s / it)
Averaged stats: lr: 0.001093  min_lr: 0.001093  loss: 4.0226 (3.7976)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:25  loss: 1.5932 (1.5932)  acc1: 69.0104 (69.0104)  acc5: 89.4531 (89.4531)  time: 9.4831  data: 9.3645  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.9804 (1.8702)  acc1: 61.5885 (62.7680)  acc5: 84.6354 (85.4400)  time: 1.1384  data: 1.0405  max mem: 21002
Test: Total time: 0:00:10 (1.1491 s / it)
* Acc@1 63.076 Acc@5 85.576 loss 1.876
Accuracy of the model on the 50000 test images: 63.1%
Max accuracy: 65.40%
Epoch: [202]  [  0/312]  eta: 1:06:40  lr: 0.001093  min_lr: 0.001093  loss: 3.5896 (3.5896)  weight_decay: 0.0500 (0.0500)  time: 12.8211  data: 12.5253  max mem: 21002
Epoch: [202]  [ 10/312]  eta: 0:08:33  lr: 0.001092  min_lr: 0.001092  loss: 3.5896 (3.6127)  weight_decay: 0.0500 (0.0500)  time: 1.7014  data: 1.1391  max mem: 21002
Epoch: [202]  [ 20/312]  eta: 0:05:12  lr: 0.001091  min_lr: 0.001091  loss: 3.6969 (3.6568)  weight_decay: 0.0500 (0.0500)  time: 0.4816  data: 0.0452  max mem: 21002
Epoch: [202]  [ 30/312]  eta: 0:03:50  lr: 0.001091  min_lr: 0.001091  loss: 3.6171 (3.6235)  weight_decay: 0.0500 (0.0500)  time: 0.3305  data: 0.0453  max mem: 21002
Epoch: [202]  [ 40/312]  eta: 0:03:16  lr: 0.001090  min_lr: 0.001090  loss: 3.9925 (3.7031)  weight_decay: 0.0500 (0.0500)  time: 0.3573  data: 0.0153  max mem: 21002
Epoch: [202]  [ 50/312]  eta: 0:03:04  lr: 0.001090  min_lr: 0.001090  loss: 4.0069 (3.7347)  weight_decay: 0.0500 (0.0500)  time: 0.5269  data: 0.1319  max mem: 21002
Epoch: [202]  [ 60/312]  eta: 0:02:44  lr: 0.001089  min_lr: 0.001089  loss: 3.8598 (3.7289)  weight_decay: 0.0500 (0.0500)  time: 0.5083  data: 0.1174  max mem: 21002
Epoch: [202]  [ 70/312]  eta: 0:02:39  lr: 0.001088  min_lr: 0.001088  loss: 3.5990 (3.7234)  weight_decay: 0.0500 (0.0500)  time: 0.5480  data: 0.0996  max mem: 21002
Epoch: [202]  [ 80/312]  eta: 0:02:25  lr: 0.001088  min_lr: 0.001088  loss: 3.7934 (3.7159)  weight_decay: 0.0500 (0.0500)  time: 0.5569  data: 0.1262  max mem: 21002
Epoch: [202]  [ 90/312]  eta: 0:02:21  lr: 0.001087  min_lr: 0.001087  loss: 3.7934 (3.7202)  weight_decay: 0.0500 (0.0500)  time: 0.5514  data: 0.0603  max mem: 21002
Epoch: [202]  [100/312]  eta: 0:02:13  lr: 0.001086  min_lr: 0.001086  loss: 3.9384 (3.7456)  weight_decay: 0.0500 (0.0500)  time: 0.6246  data: 0.0651  max mem: 21002
Epoch: [202]  [110/312]  eta: 0:02:01  lr: 0.001086  min_lr: 0.001086  loss: 4.0143 (3.7590)  weight_decay: 0.0500 (0.0500)  time: 0.4365  data: 0.0483  max mem: 21002
Epoch: [202]  [120/312]  eta: 0:01:56  lr: 0.001085  min_lr: 0.001085  loss: 3.8614 (3.7415)  weight_decay: 0.0500 (0.0500)  time: 0.5050  data: 0.0719  max mem: 21002
Epoch: [202]  [130/312]  eta: 0:01:49  lr: 0.001084  min_lr: 0.001084  loss: 3.8614 (3.7458)  weight_decay: 0.0500 (0.0500)  time: 0.6248  data: 0.1069  max mem: 21002
Epoch: [202]  [140/312]  eta: 0:01:42  lr: 0.001084  min_lr: 0.001084  loss: 4.0024 (3.7480)  weight_decay: 0.0500 (0.0500)  time: 0.5307  data: 0.0565  max mem: 21002
Epoch: [202]  [150/312]  eta: 0:01:36  lr: 0.001083  min_lr: 0.001083  loss: 3.8573 (3.7508)  weight_decay: 0.0500 (0.0500)  time: 0.5614  data: 0.0640  max mem: 21002
Epoch: [202]  [160/312]  eta: 0:01:28  lr: 0.001083  min_lr: 0.001083  loss: 4.0030 (3.7627)  weight_decay: 0.0500 (0.0500)  time: 0.4919  data: 0.0951  max mem: 21002
Epoch: [202]  [170/312]  eta: 0:01:23  lr: 0.001082  min_lr: 0.001082  loss: 4.0072 (3.7657)  weight_decay: 0.0500 (0.0500)  time: 0.5174  data: 0.0368  max mem: 21002
Epoch: [202]  [180/312]  eta: 0:01:18  lr: 0.001081  min_lr: 0.001081  loss: 3.7875 (3.7517)  weight_decay: 0.0500 (0.0500)  time: 0.6723  data: 0.0331  max mem: 21002
Epoch: [202]  [190/312]  eta: 0:01:10  lr: 0.001081  min_lr: 0.001081  loss: 3.7907 (3.7613)  weight_decay: 0.0500 (0.0500)  time: 0.4947  data: 0.0459  max mem: 21002
Epoch: [202]  [200/312]  eta: 0:01:05  lr: 0.001080  min_lr: 0.001080  loss: 3.9850 (3.7618)  weight_decay: 0.0500 (0.0500)  time: 0.4747  data: 0.0993  max mem: 21002
Epoch: [202]  [210/312]  eta: 0:00:59  lr: 0.001079  min_lr: 0.001079  loss: 3.5696 (3.7580)  weight_decay: 0.0500 (0.0500)  time: 0.5858  data: 0.1222  max mem: 21002
Epoch: [202]  [220/312]  eta: 0:00:53  lr: 0.001079  min_lr: 0.001079  loss: 3.9305 (3.7678)  weight_decay: 0.0500 (0.0500)  time: 0.5279  data: 0.0368  max mem: 21002
Epoch: [202]  [230/312]  eta: 0:00:47  lr: 0.001078  min_lr: 0.001078  loss: 3.8188 (3.7550)  weight_decay: 0.0500 (0.0500)  time: 0.5739  data: 0.0769  max mem: 21002
Epoch: [202]  [240/312]  eta: 0:00:41  lr: 0.001077  min_lr: 0.001077  loss: 3.7526 (3.7524)  weight_decay: 0.0500 (0.0500)  time: 0.5272  data: 0.1449  max mem: 21002
Epoch: [202]  [250/312]  eta: 0:00:35  lr: 0.001077  min_lr: 0.001077  loss: 3.8522 (3.7486)  weight_decay: 0.0500 (0.0500)  time: 0.4811  data: 0.0688  max mem: 21002
Epoch: [202]  [260/312]  eta: 0:00:29  lr: 0.001076  min_lr: 0.001076  loss: 3.9330 (3.7530)  weight_decay: 0.0500 (0.0500)  time: 0.6221  data: 0.0177  max mem: 21002
Epoch: [202]  [270/312]  eta: 0:00:23  lr: 0.001075  min_lr: 0.001075  loss: 3.7266 (3.7474)  weight_decay: 0.0500 (0.0500)  time: 0.4994  data: 0.0179  max mem: 21002
Epoch: [202]  [280/312]  eta: 0:00:18  lr: 0.001075  min_lr: 0.001075  loss: 3.8777 (3.7507)  weight_decay: 0.0500 (0.0500)  time: 0.4848  data: 0.0733  max mem: 21002
Epoch: [202]  [290/312]  eta: 0:00:12  lr: 0.001074  min_lr: 0.001074  loss: 3.9874 (3.7597)  weight_decay: 0.0500 (0.0500)  time: 0.5563  data: 0.1117  max mem: 21002
Epoch: [202]  [300/312]  eta: 0:00:06  lr: 0.001074  min_lr: 0.001074  loss: 3.9874 (3.7542)  weight_decay: 0.0500 (0.0500)  time: 0.4091  data: 0.0391  max mem: 21002
Epoch: [202]  [310/312]  eta: 0:00:01  lr: 0.001073  min_lr: 0.001073  loss: 3.9029 (3.7573)  weight_decay: 0.0500 (0.0500)  time: 0.3299  data: 0.0001  max mem: 21002
Epoch: [202]  [311/312]  eta: 0:00:00  lr: 0.001073  min_lr: 0.001073  loss: 3.9582 (3.7587)  weight_decay: 0.0500 (0.0500)  time: 0.2774  data: 0.0001  max mem: 21002
Epoch: [202] Total time: 0:02:52 (0.5528 s / it)
Averaged stats: lr: 0.001073  min_lr: 0.001073  loss: 3.9582 (3.7878)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.6591 (1.6591)  acc1: 68.3594 (68.3594)  acc5: 91.6667 (91.6667)  time: 8.5244  data: 8.4057  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 2.0699 (1.9536)  acc1: 62.5000 (63.3440)  acc5: 84.5052 (85.6480)  time: 1.0508  data: 0.9520  max mem: 21002
Test: Total time: 0:00:09 (1.0857 s / it)
* Acc@1 63.510 Acc@5 85.934 loss 1.957
Accuracy of the model on the 50000 test images: 63.5%
Max accuracy: 65.40%
Epoch: [203]  [  0/312]  eta: 1:00:29  lr: 0.001073  min_lr: 0.001073  loss: 3.0189 (3.0189)  weight_decay: 0.0500 (0.0500)  time: 11.6315  data: 7.7234  max mem: 21002
Epoch: [203]  [ 10/312]  eta: 0:08:54  lr: 0.001072  min_lr: 0.001072  loss: 4.0186 (3.8845)  weight_decay: 0.0500 (0.0500)  time: 1.7712  data: 0.9759  max mem: 21002
Epoch: [203]  [ 20/312]  eta: 0:05:10  lr: 0.001072  min_lr: 0.001072  loss: 4.0192 (3.9800)  weight_decay: 0.0500 (0.0500)  time: 0.5353  data: 0.1509  max mem: 21002
Epoch: [203]  [ 30/312]  eta: 0:03:49  lr: 0.001071  min_lr: 0.001071  loss: 3.9902 (3.8656)  weight_decay: 0.0500 (0.0500)  time: 0.2869  data: 0.0007  max mem: 21002
Epoch: [203]  [ 40/312]  eta: 0:03:09  lr: 0.001070  min_lr: 0.001070  loss: 3.8525 (3.8146)  weight_decay: 0.0500 (0.0500)  time: 0.3078  data: 0.0009  max mem: 21002
Epoch: [203]  [ 50/312]  eta: 0:03:01  lr: 0.001070  min_lr: 0.001070  loss: 3.8525 (3.8331)  weight_decay: 0.0500 (0.0500)  time: 0.5066  data: 0.0319  max mem: 21002
Epoch: [203]  [ 60/312]  eta: 0:02:43  lr: 0.001069  min_lr: 0.001069  loss: 3.7873 (3.8194)  weight_decay: 0.0500 (0.0500)  time: 0.5495  data: 0.0995  max mem: 21002
Epoch: [203]  [ 70/312]  eta: 0:02:34  lr: 0.001068  min_lr: 0.001068  loss: 3.8857 (3.8227)  weight_decay: 0.0500 (0.0500)  time: 0.5057  data: 0.0685  max mem: 21002
Epoch: [203]  [ 80/312]  eta: 0:02:21  lr: 0.001068  min_lr: 0.001068  loss: 3.9441 (3.8310)  weight_decay: 0.0500 (0.0500)  time: 0.5004  data: 0.0427  max mem: 21002
Epoch: [203]  [ 90/312]  eta: 0:02:18  lr: 0.001067  min_lr: 0.001067  loss: 4.0746 (3.8288)  weight_decay: 0.0500 (0.0500)  time: 0.5569  data: 0.0969  max mem: 21002
Epoch: [203]  [100/312]  eta: 0:02:11  lr: 0.001066  min_lr: 0.001066  loss: 3.7879 (3.8056)  weight_decay: 0.0500 (0.0500)  time: 0.6658  data: 0.0552  max mem: 21002
Epoch: [203]  [110/312]  eta: 0:02:01  lr: 0.001066  min_lr: 0.001066  loss: 3.6978 (3.7985)  weight_decay: 0.0500 (0.0500)  time: 0.5198  data: 0.0701  max mem: 21002
Epoch: [203]  [120/312]  eta: 0:01:56  lr: 0.001065  min_lr: 0.001065  loss: 3.6853 (3.7809)  weight_decay: 0.0500 (0.0500)  time: 0.5291  data: 0.1026  max mem: 21002
Epoch: [203]  [130/312]  eta: 0:01:50  lr: 0.001065  min_lr: 0.001065  loss: 3.8373 (3.7827)  weight_decay: 0.0500 (0.0500)  time: 0.6092  data: 0.0697  max mem: 21002
Epoch: [203]  [140/312]  eta: 0:01:41  lr: 0.001064  min_lr: 0.001064  loss: 3.9222 (3.7910)  weight_decay: 0.0500 (0.0500)  time: 0.4899  data: 0.0876  max mem: 21002
Epoch: [203]  [150/312]  eta: 0:01:35  lr: 0.001063  min_lr: 0.001063  loss: 3.9324 (3.8005)  weight_decay: 0.0500 (0.0500)  time: 0.4739  data: 0.0514  max mem: 21002
Epoch: [203]  [160/312]  eta: 0:01:27  lr: 0.001063  min_lr: 0.001063  loss: 3.9192 (3.7921)  weight_decay: 0.0500 (0.0500)  time: 0.4799  data: 0.0531  max mem: 21002
Epoch: [203]  [170/312]  eta: 0:01:23  lr: 0.001062  min_lr: 0.001062  loss: 3.7800 (3.7864)  weight_decay: 0.0500 (0.0500)  time: 0.5694  data: 0.1216  max mem: 21002
Epoch: [203]  [180/312]  eta: 0:01:16  lr: 0.001061  min_lr: 0.001061  loss: 3.7639 (3.7815)  weight_decay: 0.0500 (0.0500)  time: 0.6241  data: 0.0693  max mem: 21002
Epoch: [203]  [190/312]  eta: 0:01:10  lr: 0.001061  min_lr: 0.001061  loss: 3.7722 (3.7896)  weight_decay: 0.0500 (0.0500)  time: 0.4838  data: 0.0821  max mem: 21002
Epoch: [203]  [200/312]  eta: 0:01:04  lr: 0.001060  min_lr: 0.001060  loss: 3.6682 (3.7773)  weight_decay: 0.0500 (0.0500)  time: 0.5557  data: 0.1409  max mem: 21002
Epoch: [203]  [210/312]  eta: 0:00:58  lr: 0.001059  min_lr: 0.001059  loss: 4.0725 (3.7934)  weight_decay: 0.0500 (0.0500)  time: 0.6113  data: 0.0843  max mem: 21002
Epoch: [203]  [220/312]  eta: 0:00:52  lr: 0.001059  min_lr: 0.001059  loss: 4.0837 (3.7998)  weight_decay: 0.0500 (0.0500)  time: 0.4833  data: 0.0852  max mem: 21002
Epoch: [203]  [230/312]  eta: 0:00:46  lr: 0.001058  min_lr: 0.001058  loss: 3.8467 (3.7961)  weight_decay: 0.0500 (0.0500)  time: 0.4843  data: 0.0604  max mem: 21002
Epoch: [203]  [240/312]  eta: 0:00:40  lr: 0.001058  min_lr: 0.001058  loss: 3.8357 (3.7916)  weight_decay: 0.0500 (0.0500)  time: 0.5165  data: 0.0933  max mem: 21002
Epoch: [203]  [250/312]  eta: 0:00:35  lr: 0.001057  min_lr: 0.001057  loss: 4.0190 (3.8073)  weight_decay: 0.0500 (0.0500)  time: 0.5798  data: 0.1869  max mem: 21002
Epoch: [203]  [260/312]  eta: 0:00:29  lr: 0.001056  min_lr: 0.001056  loss: 4.1419 (3.8236)  weight_decay: 0.0500 (0.0500)  time: 0.5980  data: 0.0948  max mem: 21002
Epoch: [203]  [270/312]  eta: 0:00:23  lr: 0.001056  min_lr: 0.001056  loss: 4.0710 (3.8262)  weight_decay: 0.0500 (0.0500)  time: 0.4850  data: 0.0887  max mem: 21002
Epoch: [203]  [280/312]  eta: 0:00:18  lr: 0.001055  min_lr: 0.001055  loss: 3.7479 (3.8117)  weight_decay: 0.0500 (0.0500)  time: 0.5313  data: 0.1132  max mem: 21002
Epoch: [203]  [290/312]  eta: 0:00:12  lr: 0.001054  min_lr: 0.001054  loss: 3.4393 (3.8116)  weight_decay: 0.0500 (0.0500)  time: 0.6269  data: 0.0885  max mem: 21002
Epoch: [203]  [300/312]  eta: 0:00:06  lr: 0.001054  min_lr: 0.001054  loss: 3.8930 (3.8093)  weight_decay: 0.0500 (0.0500)  time: 0.5066  data: 0.1062  max mem: 21002
Epoch: [203]  [310/312]  eta: 0:00:01  lr: 0.001053  min_lr: 0.001053  loss: 4.0630 (3.8097)  weight_decay: 0.0500 (0.0500)  time: 0.3192  data: 0.0432  max mem: 21002
Epoch: [203]  [311/312]  eta: 0:00:00  lr: 0.001053  min_lr: 0.001053  loss: 4.0810 (3.8109)  weight_decay: 0.0500 (0.0500)  time: 0.3191  data: 0.0432  max mem: 21002
Epoch: [203] Total time: 0:02:53 (0.5556 s / it)
Averaged stats: lr: 0.001053  min_lr: 0.001053  loss: 4.0810 (3.7872)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 1.4243 (1.4243)  acc1: 73.0469 (73.0469)  acc5: 91.2760 (91.2760)  time: 8.9340  data: 8.8177  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.7382 (1.6268)  acc1: 63.2812 (65.8880)  acc5: 85.8073 (87.1520)  time: 1.0815  data: 0.9798  max mem: 21002
Test: Total time: 0:00:09 (1.1100 s / it)
* Acc@1 65.806 Acc@5 87.274 loss 1.636
Accuracy of the model on the 50000 test images: 65.8%
Max accuracy: 65.81%
Epoch: [204]  [  0/312]  eta: 1:02:38  lr: 0.001053  min_lr: 0.001053  loss: 4.1761 (4.1761)  weight_decay: 0.0500 (0.0500)  time: 12.0472  data: 11.7098  max mem: 21002
Epoch: [204]  [ 10/312]  eta: 0:07:48  lr: 0.001052  min_lr: 0.001052  loss: 4.0655 (3.9716)  weight_decay: 0.0500 (0.0500)  time: 1.5507  data: 1.0651  max mem: 21002
Epoch: [204]  [ 20/312]  eta: 0:04:59  lr: 0.001052  min_lr: 0.001052  loss: 3.9622 (3.8982)  weight_decay: 0.0500 (0.0500)  time: 0.4731  data: 0.0813  max mem: 21002
Epoch: [204]  [ 30/312]  eta: 0:03:41  lr: 0.001051  min_lr: 0.001051  loss: 3.8459 (3.7522)  weight_decay: 0.0500 (0.0500)  time: 0.3671  data: 0.0813  max mem: 21002
Epoch: [204]  [ 40/312]  eta: 0:03:17  lr: 0.001050  min_lr: 0.001050  loss: 3.9419 (3.8010)  weight_decay: 0.0500 (0.0500)  time: 0.4128  data: 0.1185  max mem: 21002
Epoch: [204]  [ 50/312]  eta: 0:03:05  lr: 0.001050  min_lr: 0.001050  loss: 3.9880 (3.8069)  weight_decay: 0.0500 (0.0500)  time: 0.5887  data: 0.2719  max mem: 21002
Epoch: [204]  [ 60/312]  eta: 0:02:41  lr: 0.001049  min_lr: 0.001049  loss: 3.9422 (3.8285)  weight_decay: 0.0500 (0.0500)  time: 0.4627  data: 0.1541  max mem: 21002
Epoch: [204]  [ 70/312]  eta: 0:02:35  lr: 0.001049  min_lr: 0.001049  loss: 3.9066 (3.8213)  weight_decay: 0.0500 (0.0500)  time: 0.4755  data: 0.1722  max mem: 21002
Epoch: [204]  [ 80/312]  eta: 0:02:32  lr: 0.001048  min_lr: 0.001048  loss: 3.9149 (3.8343)  weight_decay: 0.0500 (0.0500)  time: 0.7103  data: 0.3611  max mem: 21002
Epoch: [204]  [ 90/312]  eta: 0:02:16  lr: 0.001047  min_lr: 0.001047  loss: 3.9805 (3.8372)  weight_decay: 0.0500 (0.0500)  time: 0.5205  data: 0.1897  max mem: 21002
Epoch: [204]  [100/312]  eta: 0:02:11  lr: 0.001047  min_lr: 0.001047  loss: 3.9235 (3.8170)  weight_decay: 0.0500 (0.0500)  time: 0.4813  data: 0.1052  max mem: 21002
Epoch: [204]  [110/312]  eta: 0:01:59  lr: 0.001046  min_lr: 0.001046  loss: 3.6507 (3.7949)  weight_decay: 0.0500 (0.0500)  time: 0.4806  data: 0.1052  max mem: 21002
Epoch: [204]  [120/312]  eta: 0:01:56  lr: 0.001045  min_lr: 0.001045  loss: 3.6507 (3.7841)  weight_decay: 0.0500 (0.0500)  time: 0.5141  data: 0.1166  max mem: 21002
Epoch: [204]  [130/312]  eta: 0:01:51  lr: 0.001045  min_lr: 0.001045  loss: 3.7984 (3.7844)  weight_decay: 0.0500 (0.0500)  time: 0.7349  data: 0.1919  max mem: 21002
Epoch: [204]  [140/312]  eta: 0:01:41  lr: 0.001044  min_lr: 0.001044  loss: 4.0809 (3.8122)  weight_decay: 0.0500 (0.0500)  time: 0.5072  data: 0.0759  max mem: 21002
Epoch: [204]  [150/312]  eta: 0:01:36  lr: 0.001044  min_lr: 0.001044  loss: 4.1063 (3.8207)  weight_decay: 0.0500 (0.0500)  time: 0.4890  data: 0.0361  max mem: 21002
Epoch: [204]  [160/312]  eta: 0:01:31  lr: 0.001043  min_lr: 0.001043  loss: 3.8868 (3.8194)  weight_decay: 0.0500 (0.0500)  time: 0.6904  data: 0.0553  max mem: 21002
Epoch: [204]  [170/312]  eta: 0:01:23  lr: 0.001042  min_lr: 0.001042  loss: 3.8265 (3.8169)  weight_decay: 0.0500 (0.0500)  time: 0.4886  data: 0.0199  max mem: 21002
Epoch: [204]  [180/312]  eta: 0:01:18  lr: 0.001042  min_lr: 0.001042  loss: 3.4655 (3.8047)  weight_decay: 0.0500 (0.0500)  time: 0.4947  data: 0.0205  max mem: 21002
Epoch: [204]  [190/312]  eta: 0:01:10  lr: 0.001041  min_lr: 0.001041  loss: 3.7373 (3.8011)  weight_decay: 0.0500 (0.0500)  time: 0.4949  data: 0.0204  max mem: 21002
Epoch: [204]  [200/312]  eta: 0:01:05  lr: 0.001040  min_lr: 0.001040  loss: 3.7373 (3.7932)  weight_decay: 0.0500 (0.0500)  time: 0.4994  data: 0.0244  max mem: 21002
Epoch: [204]  [210/312]  eta: 0:01:00  lr: 0.001040  min_lr: 0.001040  loss: 3.4495 (3.7826)  weight_decay: 0.0500 (0.0500)  time: 0.7149  data: 0.0890  max mem: 21002
Epoch: [204]  [220/312]  eta: 0:00:52  lr: 0.001039  min_lr: 0.001039  loss: 3.6513 (3.7798)  weight_decay: 0.0500 (0.0500)  time: 0.5042  data: 0.0654  max mem: 21002
Epoch: [204]  [230/312]  eta: 0:00:47  lr: 0.001038  min_lr: 0.001038  loss: 3.6833 (3.7765)  weight_decay: 0.0500 (0.0500)  time: 0.4958  data: 0.0436  max mem: 21002
Epoch: [204]  [240/312]  eta: 0:00:42  lr: 0.001038  min_lr: 0.001038  loss: 3.7521 (3.7801)  weight_decay: 0.0500 (0.0500)  time: 0.6864  data: 0.0633  max mem: 21002
Epoch: [204]  [250/312]  eta: 0:00:35  lr: 0.001037  min_lr: 0.001037  loss: 3.9821 (3.7885)  weight_decay: 0.0500 (0.0500)  time: 0.4837  data: 0.0270  max mem: 21002
Epoch: [204]  [260/312]  eta: 0:00:29  lr: 0.001037  min_lr: 0.001037  loss: 4.0222 (3.7930)  weight_decay: 0.0500 (0.0500)  time: 0.4761  data: 0.0558  max mem: 21002
Epoch: [204]  [270/312]  eta: 0:00:23  lr: 0.001036  min_lr: 0.001036  loss: 4.0496 (3.7972)  weight_decay: 0.0500 (0.0500)  time: 0.4716  data: 0.0492  max mem: 21002
Epoch: [204]  [280/312]  eta: 0:00:18  lr: 0.001035  min_lr: 0.001035  loss: 3.7656 (3.7867)  weight_decay: 0.0500 (0.0500)  time: 0.4953  data: 0.0333  max mem: 21002
Epoch: [204]  [290/312]  eta: 0:00:12  lr: 0.001035  min_lr: 0.001035  loss: 3.4798 (3.7780)  weight_decay: 0.0500 (0.0500)  time: 0.6022  data: 0.0834  max mem: 21002
Epoch: [204]  [300/312]  eta: 0:00:06  lr: 0.001034  min_lr: 0.001034  loss: 3.6844 (3.7748)  weight_decay: 0.0500 (0.0500)  time: 0.3893  data: 0.0506  max mem: 21002
Epoch: [204]  [310/312]  eta: 0:00:01  lr: 0.001033  min_lr: 0.001033  loss: 3.8532 (3.7804)  weight_decay: 0.0500 (0.0500)  time: 0.2771  data: 0.0001  max mem: 21002
Epoch: [204]  [311/312]  eta: 0:00:00  lr: 0.001033  min_lr: 0.001033  loss: 3.8532 (3.7809)  weight_decay: 0.0500 (0.0500)  time: 0.2771  data: 0.0001  max mem: 21002
Epoch: [204] Total time: 0:02:52 (0.5520 s / it)
Averaged stats: lr: 0.001033  min_lr: 0.001033  loss: 3.8532 (3.7765)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.2874 (1.2874)  acc1: 74.2188 (74.2188)  acc5: 93.7500 (93.7500)  time: 8.8676  data: 8.7487  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.7578 (1.6174)  acc1: 64.1927 (66.1440)  acc5: 87.8906 (88.2720)  time: 1.0721  data: 0.9722  max mem: 21002
Test: Total time: 0:00:09 (1.1073 s / it)
* Acc@1 66.122 Acc@5 87.792 loss 1.633
Accuracy of the model on the 50000 test images: 66.1%
Max accuracy: 66.12%
Epoch: [205]  [  0/312]  eta: 1:01:01  lr: 0.001033  min_lr: 0.001033  loss: 3.8379 (3.8379)  weight_decay: 0.0500 (0.0500)  time: 11.7364  data: 8.5670  max mem: 21002
Epoch: [205]  [ 10/312]  eta: 0:08:29  lr: 0.001033  min_lr: 0.001033  loss: 4.1277 (4.0822)  weight_decay: 0.0500 (0.0500)  time: 1.6871  data: 0.9889  max mem: 21002
Epoch: [205]  [ 20/312]  eta: 0:05:11  lr: 0.001032  min_lr: 0.001032  loss: 3.9807 (3.9551)  weight_decay: 0.0500 (0.0500)  time: 0.5334  data: 0.1648  max mem: 21002
Epoch: [205]  [ 30/312]  eta: 0:03:50  lr: 0.001031  min_lr: 0.001031  loss: 3.9807 (4.0048)  weight_decay: 0.0500 (0.0500)  time: 0.3369  data: 0.0495  max mem: 21002
Epoch: [205]  [ 40/312]  eta: 0:03:08  lr: 0.001031  min_lr: 0.001031  loss: 4.0551 (3.9408)  weight_decay: 0.0500 (0.0500)  time: 0.3016  data: 0.0006  max mem: 21002
Epoch: [205]  [ 50/312]  eta: 0:03:01  lr: 0.001030  min_lr: 0.001030  loss: 3.9050 (3.9410)  weight_decay: 0.0500 (0.0500)  time: 0.5050  data: 0.0989  max mem: 21002
Epoch: [205]  [ 60/312]  eta: 0:02:38  lr: 0.001030  min_lr: 0.001030  loss: 3.9398 (3.9586)  weight_decay: 0.0500 (0.0500)  time: 0.4905  data: 0.0989  max mem: 21002
Epoch: [205]  [ 70/312]  eta: 0:02:36  lr: 0.001029  min_lr: 0.001029  loss: 4.1286 (3.9932)  weight_decay: 0.0500 (0.0500)  time: 0.5201  data: 0.1057  max mem: 21002
Epoch: [205]  [ 80/312]  eta: 0:02:28  lr: 0.001028  min_lr: 0.001028  loss: 4.0717 (3.9547)  weight_decay: 0.0500 (0.0500)  time: 0.6836  data: 0.1404  max mem: 21002
Epoch: [205]  [ 90/312]  eta: 0:02:15  lr: 0.001028  min_lr: 0.001028  loss: 3.9276 (3.9391)  weight_decay: 0.0500 (0.0500)  time: 0.4840  data: 0.0724  max mem: 21002
Epoch: [205]  [100/312]  eta: 0:02:11  lr: 0.001027  min_lr: 0.001027  loss: 3.9384 (3.9232)  weight_decay: 0.0500 (0.0500)  time: 0.5336  data: 0.1047  max mem: 21002
Epoch: [205]  [110/312]  eta: 0:02:00  lr: 0.001026  min_lr: 0.001026  loss: 3.9384 (3.9088)  weight_decay: 0.0500 (0.0500)  time: 0.5369  data: 0.1063  max mem: 21002
Epoch: [205]  [120/312]  eta: 0:01:54  lr: 0.001026  min_lr: 0.001026  loss: 4.0627 (3.9009)  weight_decay: 0.0500 (0.0500)  time: 0.4697  data: 0.0697  max mem: 21002
Epoch: [205]  [130/312]  eta: 0:01:49  lr: 0.001025  min_lr: 0.001025  loss: 4.0125 (3.8972)  weight_decay: 0.0500 (0.0500)  time: 0.6363  data: 0.1088  max mem: 21002
Epoch: [205]  [140/312]  eta: 0:01:39  lr: 0.001025  min_lr: 0.001025  loss: 4.0125 (3.9047)  weight_decay: 0.0500 (0.0500)  time: 0.4983  data: 0.0861  max mem: 21002
Epoch: [205]  [150/312]  eta: 0:01:34  lr: 0.001024  min_lr: 0.001024  loss: 3.9050 (3.8931)  weight_decay: 0.0500 (0.0500)  time: 0.4768  data: 0.0599  max mem: 21002
Epoch: [205]  [160/312]  eta: 0:01:29  lr: 0.001023  min_lr: 0.001023  loss: 3.7261 (3.8845)  weight_decay: 0.0500 (0.0500)  time: 0.6524  data: 0.1053  max mem: 21002
Epoch: [205]  [170/312]  eta: 0:01:21  lr: 0.001023  min_lr: 0.001023  loss: 3.7327 (3.8781)  weight_decay: 0.0500 (0.0500)  time: 0.5094  data: 0.0743  max mem: 21002
Epoch: [205]  [180/312]  eta: 0:01:16  lr: 0.001022  min_lr: 0.001022  loss: 3.9566 (3.8884)  weight_decay: 0.0500 (0.0500)  time: 0.5269  data: 0.0728  max mem: 21002
Epoch: [205]  [190/312]  eta: 0:01:09  lr: 0.001021  min_lr: 0.001021  loss: 3.8287 (3.8736)  weight_decay: 0.0500 (0.0500)  time: 0.5077  data: 0.0733  max mem: 21002
Epoch: [205]  [200/312]  eta: 0:01:03  lr: 0.001021  min_lr: 0.001021  loss: 3.8254 (3.8706)  weight_decay: 0.0500 (0.0500)  time: 0.4636  data: 0.0604  max mem: 21002
Epoch: [205]  [210/312]  eta: 0:00:58  lr: 0.001020  min_lr: 0.001020  loss: 3.8272 (3.8531)  weight_decay: 0.0500 (0.0500)  time: 0.6122  data: 0.1036  max mem: 21002
Epoch: [205]  [220/312]  eta: 0:00:51  lr: 0.001019  min_lr: 0.001019  loss: 3.7027 (3.8421)  weight_decay: 0.0500 (0.0500)  time: 0.4632  data: 0.0735  max mem: 21002
Epoch: [205]  [230/312]  eta: 0:00:46  lr: 0.001019  min_lr: 0.001019  loss: 3.5394 (3.8312)  weight_decay: 0.0500 (0.0500)  time: 0.5092  data: 0.1154  max mem: 21002
Epoch: [205]  [240/312]  eta: 0:00:41  lr: 0.001018  min_lr: 0.001018  loss: 3.3993 (3.8163)  weight_decay: 0.0500 (0.0500)  time: 0.6852  data: 0.1965  max mem: 21002
Epoch: [205]  [250/312]  eta: 0:00:34  lr: 0.001018  min_lr: 0.001018  loss: 3.6561 (3.8181)  weight_decay: 0.0500 (0.0500)  time: 0.4811  data: 0.0901  max mem: 21002
Epoch: [205]  [260/312]  eta: 0:00:29  lr: 0.001017  min_lr: 0.001017  loss: 3.9824 (3.8082)  weight_decay: 0.0500 (0.0500)  time: 0.4947  data: 0.0881  max mem: 21002
Epoch: [205]  [270/312]  eta: 0:00:23  lr: 0.001016  min_lr: 0.001016  loss: 4.0020 (3.8147)  weight_decay: 0.0500 (0.0500)  time: 0.5049  data: 0.0972  max mem: 21002
Epoch: [205]  [280/312]  eta: 0:00:17  lr: 0.001016  min_lr: 0.001016  loss: 4.0139 (3.8099)  weight_decay: 0.0500 (0.0500)  time: 0.4949  data: 0.1061  max mem: 21002
Epoch: [205]  [290/312]  eta: 0:00:12  lr: 0.001015  min_lr: 0.001015  loss: 3.8220 (3.8111)  weight_decay: 0.0500 (0.0500)  time: 0.6155  data: 0.1583  max mem: 21002
Epoch: [205]  [300/312]  eta: 0:00:06  lr: 0.001014  min_lr: 0.001014  loss: 4.0103 (3.8159)  weight_decay: 0.0500 (0.0500)  time: 0.4386  data: 0.0655  max mem: 21002
Epoch: [205]  [310/312]  eta: 0:00:01  lr: 0.001014  min_lr: 0.001014  loss: 4.0078 (3.8170)  weight_decay: 0.0500 (0.0500)  time: 0.2993  data: 0.0040  max mem: 21002
Epoch: [205]  [311/312]  eta: 0:00:00  lr: 0.001014  min_lr: 0.001014  loss: 4.0103 (3.8184)  weight_decay: 0.0500 (0.0500)  time: 0.2959  data: 0.0001  max mem: 21002
Epoch: [205] Total time: 0:02:50 (0.5471 s / it)
Averaged stats: lr: 0.001014  min_lr: 0.001014  loss: 4.0103 (3.7878)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 1.6281 (1.6281)  acc1: 75.0000 (75.0000)  acc5: 92.1875 (92.1875)  time: 8.1805  data: 8.0642  max mem: 21002
Test:  [8/9]  eta: 0:00:00  loss: 1.8834 (1.7929)  acc1: 63.6719 (65.5040)  acc5: 88.4115 (87.5680)  time: 0.9925  data: 0.8961  max mem: 21002
Test: Total time: 0:00:09 (1.0101 s / it)
* Acc@1 65.650 Acc@5 87.526 loss 1.809
Accuracy of the model on the 50000 test images: 65.7%
Max accuracy: 66.12%
Epoch: [206]  [  0/312]  eta: 0:56:19  lr: 0.001014  min_lr: 0.001014  loss: 3.2265 (3.2265)  weight_decay: 0.0500 (0.0500)  time: 10.8314  data: 10.5447  max mem: 21002
Epoch: [206]  [ 10/312]  eta: 0:07:22  lr: 0.001013  min_lr: 0.001013  loss: 3.4512 (3.6449)  weight_decay: 0.0500 (0.0500)  time: 1.4660  data: 1.0316  max mem: 21002
Epoch: [206]  [ 20/312]  eta: 0:05:09  lr: 0.001012  min_lr: 0.001012  loss: 3.9325 (3.8639)  weight_decay: 0.0500 (0.0500)  time: 0.5719  data: 0.1758  max mem: 21002
Epoch: [206]  [ 30/312]  eta: 0:03:48  lr: 0.001012  min_lr: 0.001012  loss: 3.9570 (3.7566)  weight_decay: 0.0500 (0.0500)  time: 0.4514  data: 0.1360  max mem: 21002
Epoch: [206]  [ 40/312]  eta: 0:03:26  lr: 0.001011  min_lr: 0.001011  loss: 3.8509 (3.8012)  weight_decay: 0.0500 (0.0500)  time: 0.4418  data: 0.1487  max mem: 21002
Epoch: [206]  [ 50/312]  eta: 0:03:15  lr: 0.001011  min_lr: 0.001011  loss: 3.9512 (3.7942)  weight_decay: 0.0500 (0.0500)  time: 0.6450  data: 0.3224  max mem: 21002
Epoch: [206]  [ 60/312]  eta: 0:02:49  lr: 0.001010  min_lr: 0.001010  loss: 3.8133 (3.7531)  weight_decay: 0.0500 (0.0500)  time: 0.4902  data: 0.1743  max mem: 21002
Epoch: [206]  [ 70/312]  eta: 0:02:43  lr: 0.001009  min_lr: 0.001009  loss: 3.8996 (3.7951)  weight_decay: 0.0500 (0.0500)  time: 0.5013  data: 0.1429  max mem: 21002
Epoch: [206]  [ 80/312]  eta: 0:02:32  lr: 0.001009  min_lr: 0.001009  loss: 3.9443 (3.7622)  weight_decay: 0.0500 (0.0500)  time: 0.6183  data: 0.2578  max mem: 21002
Epoch: [206]  [ 90/312]  eta: 0:02:19  lr: 0.001008  min_lr: 0.001008  loss: 3.9443 (3.7726)  weight_decay: 0.0500 (0.0500)  time: 0.4532  data: 0.1267  max mem: 21002
Epoch: [206]  [100/312]  eta: 0:02:14  lr: 0.001007  min_lr: 0.001007  loss: 3.8463 (3.7718)  weight_decay: 0.0500 (0.0500)  time: 0.5442  data: 0.1670  max mem: 21002
Epoch: [206]  [110/312]  eta: 0:02:01  lr: 0.001007  min_lr: 0.001007  loss: 3.7982 (3.7616)  weight_decay: 0.0500 (0.0500)  time: 0.4922  data: 0.1559  max mem: 21002
Epoch: [206]  [120/312]  eta: 0:01:56  lr: 0.001006  min_lr: 0.001006  loss: 3.7982 (3.7445)  weight_decay: 0.0500 (0.0500)  time: 0.4610  data: 0.1366  max mem: 21002
Epoch: [206]  [130/312]  eta: 0:01:51  lr: 0.001006  min_lr: 0.001006  loss: 3.5819 (3.7301)  weight_decay: 0.0500 (0.0500)  time: 0.6569  data: 0.3006  max mem: 21002
Epoch: [206]  [140/312]  eta: 0:01:41  lr: 0.001005  min_lr: 0.001005  loss: 3.5763 (3.7165)  weight_decay: 0.0500 (0.0500)  time: 0.4809  data: 0.1647  max mem: 21002
Epoch: [206]  [150/312]  eta: 0:01:36  lr: 0.001004  min_lr: 0.001004  loss: 3.9070 (3.7390)  weight_decay: 0.0500 (0.0500)  time: 0.4785  data: 0.1748  max mem: 21002
Epoch: [206]  [160/312]  eta: 0:01:30  lr: 0.001004  min_lr: 0.001004  loss: 4.0380 (3.7454)  weight_decay: 0.0500 (0.0500)  time: 0.6496  data: 0.3462  max mem: 21002
Epoch: [206]  [170/312]  eta: 0:01:22  lr: 0.001003  min_lr: 0.001003  loss: 3.9424 (3.7684)  weight_decay: 0.0500 (0.0500)  time: 0.4791  data: 0.1721  max mem: 21002
Epoch: [206]  [180/312]  eta: 0:01:17  lr: 0.001002  min_lr: 0.001002  loss: 4.0470 (3.7722)  weight_decay: 0.0500 (0.0500)  time: 0.5269  data: 0.1793  max mem: 21002
Epoch: [206]  [190/312]  eta: 0:01:09  lr: 0.001002  min_lr: 0.001002  loss: 3.7722 (3.7638)  weight_decay: 0.0500 (0.0500)  time: 0.5071  data: 0.1792  max mem: 21002
Epoch: [206]  [200/312]  eta: 0:01:04  lr: 0.001001  min_lr: 0.001001  loss: 3.4899 (3.7476)  weight_decay: 0.0500 (0.0500)  time: 0.4941  data: 0.1525  max mem: 21002
Epoch: [206]  [210/312]  eta: 0:00:59  lr: 0.001001  min_lr: 0.001001  loss: 3.5198 (3.7481)  weight_decay: 0.0500 (0.0500)  time: 0.6847  data: 0.3021  max mem: 21002
Epoch: [206]  [220/312]  eta: 0:00:52  lr: 0.001000  min_lr: 0.001000  loss: 3.6843 (3.7376)  weight_decay: 0.0500 (0.0500)  time: 0.4783  data: 0.1502  max mem: 21002
Epoch: [206]  [230/312]  eta: 0:00:47  lr: 0.000999  min_lr: 0.000999  loss: 3.7732 (3.7441)  weight_decay: 0.0500 (0.0500)  time: 0.4885  data: 0.1415  max mem: 21002
Epoch: [206]  [240/312]  eta: 0:00:41  lr: 0.000999  min_lr: 0.000999  loss: 4.1613 (3.7598)  weight_decay: 0.0500 (0.0500)  time: 0.6605  data: 0.2835  max mem: 21002
Epoch: [206]  [250/312]  eta: 0:00:35  lr: 0.000998  min_lr: 0.000998  loss: 4.0201 (3.7604)  weight_decay: 0.0500 (0.0500)  time: 0.4784  data: 0.1426  max mem: 21002
Epoch: [206]  [260/312]  eta: 0:00:29  lr: 0.000998  min_lr: 0.000998  loss: 3.9277 (3.7557)  weight_decay: 0.0500 (0.0500)  time: 0.5009  data: 0.1358  max mem: 21002
Epoch: [206]  [270/312]  eta: 0:00:23  lr: 0.000997  min_lr: 0.000997  loss: 3.8494 (3.7570)  weight_decay: 0.0500 (0.0500)  time: 0.4822  data: 0.1366  max mem: 21002
Epoch: [206]  [280/312]  eta: 0:00:18  lr: 0.000996  min_lr: 0.000996  loss: 3.7110 (3.7537)  weight_decay: 0.0500 (0.0500)  time: 0.4830  data: 0.1188  max mem: 21002
Epoch: [206]  [290/312]  eta: 0:00:12  lr: 0.000996  min_lr: 0.000996  loss: 3.8663 (3.7569)  weight_decay: 0.0500 (0.0500)  time: 0.6157  data: 0.2012  max mem: 21002
Epoch: [206]  [300/312]  eta: 0:00:06  lr: 0.000995  min_lr: 0.000995  loss: 3.6774 (3.7471)  weight_decay: 0.0500 (0.0500)  time: 0.4163  data: 0.0836  max mem: 21002
Epoch: [206]  [310/312]  eta: 0:00:01  lr: 0.000994  min_lr: 0.000994  loss: 3.3919 (3.7427)  weight_decay: 0.0500 (0.0500)  time: 0.2776  data: 0.0001  max mem: 21002
Epoch: [206]  [311/312]  eta: 0:00:00  lr: 0.000994  min_lr: 0.000994  loss: 3.3919 (3.7419)  weight_decay: 0.0500 (0.0500)  time: 0.2776  data: 0.0001  max mem: 21002
Epoch: [206] Total time: 0:02:51 (0.5493 s / it)
Averaged stats: lr: 0.000994  min_lr: 0.000994  loss: 3.3919 (3.7539)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:21  loss: 1.4499 (1.4499)  acc1: 71.8750 (71.8750)  acc5: 91.0156 (91.0156)  time: 9.1101  data: 8.9913  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.8635 (1.6808)  acc1: 62.8906 (64.5760)  acc5: 84.5052 (86.2560)  time: 1.0989  data: 0.9991  max mem: 21002
Test: Total time: 0:00:10 (1.1331 s / it)
* Acc@1 64.860 Acc@5 86.818 loss 1.694
Accuracy of the model on the 50000 test images: 64.9%
Max accuracy: 66.12%
Epoch: [207]  [  0/312]  eta: 1:01:38  lr: 0.000994  min_lr: 0.000994  loss: 3.3525 (3.3525)  weight_decay: 0.0500 (0.0500)  time: 11.8538  data: 9.5252  max mem: 21002
Epoch: [207]  [ 10/312]  eta: 0:08:19  lr: 0.000994  min_lr: 0.000994  loss: 3.8570 (3.8220)  weight_decay: 0.0500 (0.0500)  time: 1.6529  data: 1.0284  max mem: 21002
Epoch: [207]  [ 20/312]  eta: 0:05:04  lr: 0.000993  min_lr: 0.000993  loss: 3.8209 (3.7074)  weight_decay: 0.0500 (0.0500)  time: 0.5030  data: 0.0933  max mem: 21002
Epoch: [207]  [ 30/312]  eta: 0:03:45  lr: 0.000992  min_lr: 0.000992  loss: 3.8209 (3.7227)  weight_decay: 0.0500 (0.0500)  time: 0.3300  data: 0.0043  max mem: 21002
Epoch: [207]  [ 40/312]  eta: 0:03:10  lr: 0.000992  min_lr: 0.000992  loss: 3.6651 (3.6765)  weight_decay: 0.0500 (0.0500)  time: 0.3422  data: 0.0014  max mem: 21002
Epoch: [207]  [ 50/312]  eta: 0:03:04  lr: 0.000991  min_lr: 0.000991  loss: 3.5273 (3.6856)  weight_decay: 0.0500 (0.0500)  time: 0.5547  data: 0.0944  max mem: 21002
Epoch: [207]  [ 60/312]  eta: 0:02:44  lr: 0.000991  min_lr: 0.000991  loss: 3.5267 (3.6789)  weight_decay: 0.0500 (0.0500)  time: 0.5522  data: 0.1308  max mem: 21002
Epoch: [207]  [ 70/312]  eta: 0:02:35  lr: 0.000990  min_lr: 0.000990  loss: 3.8043 (3.7115)  weight_decay: 0.0500 (0.0500)  time: 0.4893  data: 0.0876  max mem: 21002
Epoch: [207]  [ 80/312]  eta: 0:02:26  lr: 0.000989  min_lr: 0.000989  loss: 3.7252 (3.6930)  weight_decay: 0.0500 (0.0500)  time: 0.5751  data: 0.0944  max mem: 21002
Epoch: [207]  [ 90/312]  eta: 0:02:15  lr: 0.000989  min_lr: 0.000989  loss: 3.5425 (3.6711)  weight_decay: 0.0500 (0.0500)  time: 0.4854  data: 0.1043  max mem: 21002
Epoch: [207]  [100/312]  eta: 0:02:10  lr: 0.000988  min_lr: 0.000988  loss: 3.8612 (3.6936)  weight_decay: 0.0500 (0.0500)  time: 0.5382  data: 0.1287  max mem: 21002
Epoch: [207]  [110/312]  eta: 0:01:58  lr: 0.000987  min_lr: 0.000987  loss: 3.8303 (3.6881)  weight_decay: 0.0500 (0.0500)  time: 0.4792  data: 0.0692  max mem: 21002
Epoch: [207]  [120/312]  eta: 0:01:53  lr: 0.000987  min_lr: 0.000987  loss: 3.7818 (3.6990)  weight_decay: 0.0500 (0.0500)  time: 0.4820  data: 0.0706  max mem: 21002
Epoch: [207]  [130/312]  eta: 0:01:49  lr: 0.000986  min_lr: 0.000986  loss: 3.7818 (3.6941)  weight_decay: 0.0500 (0.0500)  time: 0.6964  data: 0.1685  max mem: 21002
Epoch: [207]  [140/312]  eta: 0:01:39  lr: 0.000986  min_lr: 0.000986  loss: 3.4686 (3.6647)  weight_decay: 0.0500 (0.0500)  time: 0.5016  data: 0.0987  max mem: 21002
Epoch: [207]  [150/312]  eta: 0:01:34  lr: 0.000985  min_lr: 0.000985  loss: 3.4522 (3.6693)  weight_decay: 0.0500 (0.0500)  time: 0.4648  data: 0.0524  max mem: 21002
Epoch: [207]  [160/312]  eta: 0:01:28  lr: 0.000984  min_lr: 0.000984  loss: 3.8484 (3.6750)  weight_decay: 0.0500 (0.0500)  time: 0.6057  data: 0.0625  max mem: 21002
Epoch: [207]  [170/312]  eta: 0:01:21  lr: 0.000984  min_lr: 0.000984  loss: 3.9522 (3.6833)  weight_decay: 0.0500 (0.0500)  time: 0.4855  data: 0.0687  max mem: 21002
Epoch: [207]  [180/312]  eta: 0:01:16  lr: 0.000983  min_lr: 0.000983  loss: 3.9522 (3.6740)  weight_decay: 0.0500 (0.0500)  time: 0.5278  data: 0.1167  max mem: 21002
Epoch: [207]  [190/312]  eta: 0:01:08  lr: 0.000982  min_lr: 0.000982  loss: 3.7875 (3.6716)  weight_decay: 0.0500 (0.0500)  time: 0.4690  data: 0.0587  max mem: 21002
Epoch: [207]  [200/312]  eta: 0:01:03  lr: 0.000982  min_lr: 0.000982  loss: 3.9597 (3.6855)  weight_decay: 0.0500 (0.0500)  time: 0.5163  data: 0.0728  max mem: 21002
Epoch: [207]  [210/312]  eta: 0:00:58  lr: 0.000981  min_lr: 0.000981  loss: 3.8682 (3.6816)  weight_decay: 0.0500 (0.0500)  time: 0.7012  data: 0.1180  max mem: 21002
Epoch: [207]  [220/312]  eta: 0:00:51  lr: 0.000981  min_lr: 0.000981  loss: 3.6750 (3.6724)  weight_decay: 0.0500 (0.0500)  time: 0.4713  data: 0.0459  max mem: 21002
Epoch: [207]  [230/312]  eta: 0:00:46  lr: 0.000980  min_lr: 0.000980  loss: 3.6789 (3.6809)  weight_decay: 0.0500 (0.0500)  time: 0.4971  data: 0.0520  max mem: 21002
Epoch: [207]  [240/312]  eta: 0:00:40  lr: 0.000979  min_lr: 0.000979  loss: 3.8968 (3.6854)  weight_decay: 0.0500 (0.0500)  time: 0.6138  data: 0.0520  max mem: 21002
Epoch: [207]  [250/312]  eta: 0:00:34  lr: 0.000979  min_lr: 0.000979  loss: 3.6966 (3.6814)  weight_decay: 0.0500 (0.0500)  time: 0.4826  data: 0.0532  max mem: 21002
Epoch: [207]  [260/312]  eta: 0:00:29  lr: 0.000978  min_lr: 0.000978  loss: 3.6765 (3.6834)  weight_decay: 0.0500 (0.0500)  time: 0.5819  data: 0.1038  max mem: 21002
Epoch: [207]  [270/312]  eta: 0:00:23  lr: 0.000978  min_lr: 0.000978  loss: 3.8241 (3.6899)  weight_decay: 0.0500 (0.0500)  time: 0.5062  data: 0.0512  max mem: 21002
Epoch: [207]  [280/312]  eta: 0:00:17  lr: 0.000977  min_lr: 0.000977  loss: 3.7642 (3.6903)  weight_decay: 0.0500 (0.0500)  time: 0.4766  data: 0.0553  max mem: 21002
Epoch: [207]  [290/312]  eta: 0:00:12  lr: 0.000976  min_lr: 0.000976  loss: 4.1694 (3.7093)  weight_decay: 0.0500 (0.0500)  time: 0.6100  data: 0.0872  max mem: 21002
Epoch: [207]  [300/312]  eta: 0:00:06  lr: 0.000976  min_lr: 0.000976  loss: 4.2542 (3.7227)  weight_decay: 0.0500 (0.0500)  time: 0.4267  data: 0.0404  max mem: 21002
Epoch: [207]  [310/312]  eta: 0:00:01  lr: 0.000975  min_lr: 0.000975  loss: 4.0906 (3.7269)  weight_decay: 0.0500 (0.0500)  time: 0.2852  data: 0.0083  max mem: 21002
Epoch: [207]  [311/312]  eta: 0:00:00  lr: 0.000975  min_lr: 0.000975  loss: 4.0906 (3.7296)  weight_decay: 0.0500 (0.0500)  time: 0.2851  data: 0.0083  max mem: 21002
Epoch: [207] Total time: 0:02:50 (0.5461 s / it)
Averaged stats: lr: 0.000975  min_lr: 0.000975  loss: 4.0906 (3.7689)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 1.5548 (1.5548)  acc1: 75.3906 (75.3906)  acc5: 92.8385 (92.8385)  time: 8.0427  data: 7.9246  max mem: 21002
Test:  [8/9]  eta: 0:00:00  loss: 1.9774 (1.8569)  acc1: 63.8021 (65.3440)  acc5: 85.9375 (86.9600)  time: 0.9953  data: 0.8918  max mem: 21002
Test: Total time: 0:00:09 (1.0050 s / it)
* Acc@1 65.098 Acc@5 87.052 loss 1.870
Accuracy of the model on the 50000 test images: 65.1%
Max accuracy: 66.12%
Epoch: [208]  [  0/312]  eta: 1:01:58  lr: 0.000975  min_lr: 0.000975  loss: 3.7652 (3.7652)  weight_decay: 0.0500 (0.0500)  time: 11.9175  data: 10.6560  max mem: 21002
Epoch: [208]  [ 10/312]  eta: 0:08:19  lr: 0.000974  min_lr: 0.000974  loss: 3.5572 (3.3909)  weight_decay: 0.0500 (0.0500)  time: 1.6532  data: 1.0261  max mem: 21002
Epoch: [208]  [ 20/312]  eta: 0:05:12  lr: 0.000974  min_lr: 0.000974  loss: 3.7883 (3.6524)  weight_decay: 0.0500 (0.0500)  time: 0.5280  data: 0.0691  max mem: 21002
Epoch: [208]  [ 30/312]  eta: 0:03:50  lr: 0.000973  min_lr: 0.000973  loss: 3.7851 (3.6308)  weight_decay: 0.0500 (0.0500)  time: 0.3599  data: 0.0378  max mem: 21002
Epoch: [208]  [ 40/312]  eta: 0:03:18  lr: 0.000972  min_lr: 0.000972  loss: 3.7563 (3.6801)  weight_decay: 0.0500 (0.0500)  time: 0.3688  data: 0.0009  max mem: 21002
Epoch: [208]  [ 50/312]  eta: 0:03:05  lr: 0.000972  min_lr: 0.000972  loss: 3.7907 (3.7028)  weight_decay: 0.0500 (0.0500)  time: 0.5386  data: 0.1089  max mem: 21002
Epoch: [208]  [ 60/312]  eta: 0:02:41  lr: 0.000971  min_lr: 0.000971  loss: 3.8066 (3.7201)  weight_decay: 0.0500 (0.0500)  time: 0.4613  data: 0.1088  max mem: 21002
Epoch: [208]  [ 70/312]  eta: 0:02:34  lr: 0.000971  min_lr: 0.000971  loss: 3.8604 (3.7318)  weight_decay: 0.0500 (0.0500)  time: 0.4611  data: 0.1045  max mem: 21002
Epoch: [208]  [ 80/312]  eta: 0:02:25  lr: 0.000970  min_lr: 0.000970  loss: 3.9537 (3.7444)  weight_decay: 0.0500 (0.0500)  time: 0.5947  data: 0.2150  max mem: 21002
Epoch: [208]  [ 90/312]  eta: 0:02:15  lr: 0.000969  min_lr: 0.000969  loss: 4.0998 (3.7879)  weight_decay: 0.0500 (0.0500)  time: 0.5077  data: 0.1113  max mem: 21002
Epoch: [208]  [100/312]  eta: 0:02:10  lr: 0.000969  min_lr: 0.000969  loss: 4.0998 (3.7920)  weight_decay: 0.0500 (0.0500)  time: 0.5664  data: 0.0810  max mem: 21002
Epoch: [208]  [110/312]  eta: 0:01:58  lr: 0.000968  min_lr: 0.000968  loss: 3.5665 (3.7583)  weight_decay: 0.0500 (0.0500)  time: 0.4815  data: 0.0809  max mem: 21002
Epoch: [208]  [120/312]  eta: 0:01:54  lr: 0.000968  min_lr: 0.000968  loss: 3.7687 (3.7602)  weight_decay: 0.0500 (0.0500)  time: 0.4878  data: 0.0918  max mem: 21002
Epoch: [208]  [130/312]  eta: 0:01:49  lr: 0.000967  min_lr: 0.000967  loss: 3.7892 (3.7658)  weight_decay: 0.0500 (0.0500)  time: 0.6897  data: 0.1984  max mem: 21002
Epoch: [208]  [140/312]  eta: 0:01:39  lr: 0.000966  min_lr: 0.000966  loss: 3.7842 (3.7704)  weight_decay: 0.0500 (0.0500)  time: 0.4895  data: 0.1074  max mem: 21002
Epoch: [208]  [150/312]  eta: 0:01:35  lr: 0.000966  min_lr: 0.000966  loss: 3.6365 (3.7631)  weight_decay: 0.0500 (0.0500)  time: 0.4862  data: 0.1123  max mem: 21002
Epoch: [208]  [160/312]  eta: 0:01:28  lr: 0.000965  min_lr: 0.000965  loss: 3.7412 (3.7557)  weight_decay: 0.0500 (0.0500)  time: 0.6016  data: 0.2299  max mem: 21002
Epoch: [208]  [170/312]  eta: 0:01:21  lr: 0.000964  min_lr: 0.000964  loss: 3.9019 (3.7769)  weight_decay: 0.0500 (0.0500)  time: 0.4858  data: 0.1183  max mem: 21002
Epoch: [208]  [180/312]  eta: 0:01:16  lr: 0.000964  min_lr: 0.000964  loss: 3.9879 (3.7788)  weight_decay: 0.0500 (0.0500)  time: 0.5648  data: 0.1080  max mem: 21002
Epoch: [208]  [190/312]  eta: 0:01:09  lr: 0.000963  min_lr: 0.000963  loss: 3.7782 (3.7740)  weight_decay: 0.0500 (0.0500)  time: 0.4824  data: 0.1081  max mem: 21002
Epoch: [208]  [200/312]  eta: 0:01:03  lr: 0.000963  min_lr: 0.000963  loss: 3.8715 (3.7766)  weight_decay: 0.0500 (0.0500)  time: 0.4807  data: 0.0482  max mem: 21002
Epoch: [208]  [210/312]  eta: 0:00:58  lr: 0.000962  min_lr: 0.000962  loss: 3.9257 (3.7786)  weight_decay: 0.0500 (0.0500)  time: 0.6789  data: 0.1567  max mem: 21002
Epoch: [208]  [220/312]  eta: 0:00:51  lr: 0.000961  min_lr: 0.000961  loss: 3.8528 (3.7751)  weight_decay: 0.0500 (0.0500)  time: 0.4850  data: 0.1093  max mem: 21002
Epoch: [208]  [230/312]  eta: 0:00:46  lr: 0.000961  min_lr: 0.000961  loss: 3.7235 (3.7618)  weight_decay: 0.0500 (0.0500)  time: 0.4991  data: 0.1087  max mem: 21002
Epoch: [208]  [240/312]  eta: 0:00:40  lr: 0.000960  min_lr: 0.000960  loss: 3.8018 (3.7615)  weight_decay: 0.0500 (0.0500)  time: 0.5985  data: 0.2081  max mem: 21002
Epoch: [208]  [250/312]  eta: 0:00:34  lr: 0.000960  min_lr: 0.000960  loss: 3.9350 (3.7688)  weight_decay: 0.0500 (0.0500)  time: 0.4902  data: 0.1001  max mem: 21002
Epoch: [208]  [260/312]  eta: 0:00:29  lr: 0.000959  min_lr: 0.000959  loss: 3.7846 (3.7588)  weight_decay: 0.0500 (0.0500)  time: 0.5936  data: 0.0917  max mem: 21002
Epoch: [208]  [270/312]  eta: 0:00:23  lr: 0.000958  min_lr: 0.000958  loss: 3.4471 (3.7540)  weight_decay: 0.0500 (0.0500)  time: 0.4900  data: 0.0918  max mem: 21002
Epoch: [208]  [280/312]  eta: 0:00:18  lr: 0.000958  min_lr: 0.000958  loss: 3.7248 (3.7515)  weight_decay: 0.0500 (0.0500)  time: 0.5048  data: 0.0835  max mem: 21002
Epoch: [208]  [290/312]  eta: 0:00:12  lr: 0.000957  min_lr: 0.000957  loss: 4.0017 (3.7606)  weight_decay: 0.0500 (0.0500)  time: 0.6419  data: 0.1398  max mem: 21002
Epoch: [208]  [300/312]  eta: 0:00:06  lr: 0.000956  min_lr: 0.000956  loss: 3.9113 (3.7608)  weight_decay: 0.0500 (0.0500)  time: 0.4199  data: 0.0567  max mem: 21002
Epoch: [208]  [310/312]  eta: 0:00:01  lr: 0.000956  min_lr: 0.000956  loss: 3.7783 (3.7603)  weight_decay: 0.0500 (0.0500)  time: 0.2772  data: 0.0001  max mem: 21002
Epoch: [208]  [311/312]  eta: 0:00:00  lr: 0.000956  min_lr: 0.000956  loss: 3.7045 (3.7566)  weight_decay: 0.0500 (0.0500)  time: 0.2772  data: 0.0001  max mem: 21002
Epoch: [208] Total time: 0:02:50 (0.5481 s / it)
Averaged stats: lr: 0.000956  min_lr: 0.000956  loss: 3.7045 (3.7665)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:21  loss: 1.4237 (1.4237)  acc1: 74.7396 (74.7396)  acc5: 92.4479 (92.4479)  time: 9.0625  data: 8.9442  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.8327 (1.7396)  acc1: 64.7135 (65.9200)  acc5: 87.6302 (87.2160)  time: 1.0906  data: 0.9939  max mem: 21002
Test: Total time: 0:00:09 (1.1021 s / it)
* Acc@1 65.494 Acc@5 87.204 loss 1.750
Accuracy of the model on the 50000 test images: 65.5%
Max accuracy: 66.12%
Epoch: [209]  [  0/312]  eta: 0:57:57  lr: 0.000956  min_lr: 0.000956  loss: 4.1161 (4.1161)  weight_decay: 0.0500 (0.0500)  time: 11.1442  data: 9.2657  max mem: 21002
Epoch: [209]  [ 10/312]  eta: 0:08:30  lr: 0.000955  min_lr: 0.000955  loss: 3.9729 (3.7712)  weight_decay: 0.0500 (0.0500)  time: 1.6916  data: 1.0783  max mem: 21002
Epoch: [209]  [ 20/312]  eta: 0:05:00  lr: 0.000955  min_lr: 0.000955  loss: 3.7419 (3.6669)  weight_decay: 0.0500 (0.0500)  time: 0.5238  data: 0.1301  max mem: 21002
Epoch: [209]  [ 30/312]  eta: 0:03:43  lr: 0.000954  min_lr: 0.000954  loss: 3.7017 (3.6903)  weight_decay: 0.0500 (0.0500)  time: 0.2968  data: 0.0007  max mem: 21002
Epoch: [209]  [ 40/312]  eta: 0:03:08  lr: 0.000953  min_lr: 0.000953  loss: 3.8939 (3.7004)  weight_decay: 0.0500 (0.0500)  time: 0.3374  data: 0.0120  max mem: 21002
Epoch: [209]  [ 50/312]  eta: 0:03:00  lr: 0.000953  min_lr: 0.000953  loss: 3.6971 (3.6764)  weight_decay: 0.0500 (0.0500)  time: 0.5259  data: 0.2027  max mem: 21002
Epoch: [209]  [ 60/312]  eta: 0:02:36  lr: 0.000952  min_lr: 0.000952  loss: 3.8147 (3.7254)  weight_decay: 0.0500 (0.0500)  time: 0.4797  data: 0.1915  max mem: 21002
Epoch: [209]  [ 70/312]  eta: 0:02:31  lr: 0.000951  min_lr: 0.000951  loss: 3.8180 (3.6902)  weight_decay: 0.0500 (0.0500)  time: 0.4630  data: 0.1665  max mem: 21002
Epoch: [209]  [ 80/312]  eta: 0:02:25  lr: 0.000951  min_lr: 0.000951  loss: 3.4533 (3.6522)  weight_decay: 0.0500 (0.0500)  time: 0.6504  data: 0.3343  max mem: 21002
Epoch: [209]  [ 90/312]  eta: 0:02:11  lr: 0.000950  min_lr: 0.000950  loss: 3.6819 (3.6702)  weight_decay: 0.0500 (0.0500)  time: 0.4802  data: 0.1685  max mem: 21002
Epoch: [209]  [100/312]  eta: 0:02:06  lr: 0.000950  min_lr: 0.000950  loss: 3.9346 (3.6854)  weight_decay: 0.0500 (0.0500)  time: 0.4730  data: 0.1675  max mem: 21002
Epoch: [209]  [110/312]  eta: 0:01:56  lr: 0.000949  min_lr: 0.000949  loss: 3.7848 (3.6826)  weight_decay: 0.0500 (0.0500)  time: 0.5065  data: 0.1676  max mem: 21002
Epoch: [209]  [120/312]  eta: 0:01:52  lr: 0.000948  min_lr: 0.000948  loss: 3.8646 (3.6931)  weight_decay: 0.0500 (0.0500)  time: 0.5246  data: 0.1504  max mem: 21002
Epoch: [209]  [130/312]  eta: 0:01:47  lr: 0.000948  min_lr: 0.000948  loss: 4.0140 (3.6913)  weight_decay: 0.0500 (0.0500)  time: 0.6797  data: 0.3443  max mem: 21002
Epoch: [209]  [140/312]  eta: 0:01:38  lr: 0.000947  min_lr: 0.000947  loss: 3.9949 (3.7032)  weight_decay: 0.0500 (0.0500)  time: 0.4832  data: 0.1946  max mem: 21002
Epoch: [209]  [150/312]  eta: 0:01:33  lr: 0.000947  min_lr: 0.000947  loss: 3.9949 (3.7031)  weight_decay: 0.0500 (0.0500)  time: 0.4794  data: 0.1898  max mem: 21002
Epoch: [209]  [160/312]  eta: 0:01:28  lr: 0.000946  min_lr: 0.000946  loss: 3.7173 (3.7087)  weight_decay: 0.0500 (0.0500)  time: 0.6646  data: 0.3647  max mem: 21002
Epoch: [209]  [170/312]  eta: 0:01:20  lr: 0.000945  min_lr: 0.000945  loss: 3.9387 (3.7131)  weight_decay: 0.0500 (0.0500)  time: 0.5049  data: 0.1756  max mem: 21002
Epoch: [209]  [180/312]  eta: 0:01:15  lr: 0.000945  min_lr: 0.000945  loss: 4.0223 (3.7317)  weight_decay: 0.0500 (0.0500)  time: 0.4920  data: 0.1738  max mem: 21002
Epoch: [209]  [190/312]  eta: 0:01:08  lr: 0.000944  min_lr: 0.000944  loss: 3.9526 (3.7178)  weight_decay: 0.0500 (0.0500)  time: 0.4732  data: 0.1739  max mem: 21002
Epoch: [209]  [200/312]  eta: 0:01:03  lr: 0.000944  min_lr: 0.000944  loss: 3.6424 (3.7223)  weight_decay: 0.0500 (0.0500)  time: 0.4914  data: 0.1921  max mem: 21002
Epoch: [209]  [210/312]  eta: 0:00:58  lr: 0.000943  min_lr: 0.000943  loss: 3.8955 (3.7326)  weight_decay: 0.0500 (0.0500)  time: 0.6956  data: 0.4081  max mem: 21002
Epoch: [209]  [220/312]  eta: 0:00:51  lr: 0.000942  min_lr: 0.000942  loss: 4.0074 (3.7453)  weight_decay: 0.0500 (0.0500)  time: 0.5026  data: 0.2168  max mem: 21002
Epoch: [209]  [230/312]  eta: 0:00:46  lr: 0.000942  min_lr: 0.000942  loss: 3.7292 (3.7304)  weight_decay: 0.0500 (0.0500)  time: 0.4655  data: 0.1755  max mem: 21002
Epoch: [209]  [240/312]  eta: 0:00:40  lr: 0.000941  min_lr: 0.000941  loss: 3.4930 (3.7209)  weight_decay: 0.0500 (0.0500)  time: 0.6783  data: 0.3879  max mem: 21002
Epoch: [209]  [250/312]  eta: 0:00:34  lr: 0.000940  min_lr: 0.000940  loss: 3.7636 (3.7242)  weight_decay: 0.0500 (0.0500)  time: 0.5011  data: 0.2138  max mem: 21002
Epoch: [209]  [260/312]  eta: 0:00:29  lr: 0.000940  min_lr: 0.000940  loss: 3.7196 (3.7209)  weight_decay: 0.0500 (0.0500)  time: 0.4828  data: 0.1910  max mem: 21002
Epoch: [209]  [270/312]  eta: 0:00:23  lr: 0.000939  min_lr: 0.000939  loss: 3.4626 (3.7195)  weight_decay: 0.0500 (0.0500)  time: 0.4819  data: 0.1902  max mem: 21002
Epoch: [209]  [280/312]  eta: 0:00:17  lr: 0.000939  min_lr: 0.000939  loss: 3.7520 (3.7181)  weight_decay: 0.0500 (0.0500)  time: 0.4916  data: 0.1963  max mem: 21002
Epoch: [209]  [290/312]  eta: 0:00:12  lr: 0.000938  min_lr: 0.000938  loss: 3.7520 (3.7181)  weight_decay: 0.0500 (0.0500)  time: 0.6598  data: 0.3655  max mem: 21002
Epoch: [209]  [300/312]  eta: 0:00:06  lr: 0.000937  min_lr: 0.000937  loss: 3.6942 (3.7215)  weight_decay: 0.0500 (0.0500)  time: 0.4505  data: 0.1696  max mem: 21002
Epoch: [209]  [310/312]  eta: 0:00:01  lr: 0.000937  min_lr: 0.000937  loss: 3.6942 (3.7214)  weight_decay: 0.0500 (0.0500)  time: 0.2955  data: 0.0187  max mem: 21002
Epoch: [209]  [311/312]  eta: 0:00:00  lr: 0.000937  min_lr: 0.000937  loss: 3.6216 (3.7177)  weight_decay: 0.0500 (0.0500)  time: 0.2954  data: 0.0187  max mem: 21002
Epoch: [209] Total time: 0:02:50 (0.5457 s / it)
Averaged stats: lr: 0.000937  min_lr: 0.000937  loss: 3.6216 (3.7693)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 1.4554 (1.4554)  acc1: 75.0000 (75.0000)  acc5: 91.2760 (91.2760)  time: 8.9301  data: 8.8114  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.7152 (1.5846)  acc1: 65.8854 (66.9920)  acc5: 88.1510 (87.9200)  time: 1.0906  data: 0.9791  max mem: 21002
Test: Total time: 0:00:10 (1.1193 s / it)
* Acc@1 66.788 Acc@5 88.002 loss 1.592
Accuracy of the model on the 50000 test images: 66.8%
Max accuracy: 66.79%
Epoch: [210]  [  0/312]  eta: 0:59:52  lr: 0.000937  min_lr: 0.000937  loss: 4.6661 (4.6661)  weight_decay: 0.0500 (0.0500)  time: 11.5156  data: 10.2976  max mem: 21002
Epoch: [210]  [ 10/312]  eta: 0:08:29  lr: 0.000936  min_lr: 0.000936  loss: 3.8867 (3.8736)  weight_decay: 0.0500 (0.0500)  time: 1.6878  data: 1.0318  max mem: 21002
Epoch: [210]  [ 20/312]  eta: 0:05:25  lr: 0.000935  min_lr: 0.000935  loss: 3.7411 (3.8684)  weight_decay: 0.0500 (0.0500)  time: 0.5956  data: 0.0822  max mem: 21002
Epoch: [210]  [ 30/312]  eta: 0:03:59  lr: 0.000935  min_lr: 0.000935  loss: 3.8193 (3.7719)  weight_decay: 0.0500 (0.0500)  time: 0.3897  data: 0.0300  max mem: 21002
Epoch: [210]  [ 40/312]  eta: 0:03:14  lr: 0.000934  min_lr: 0.000934  loss: 3.8193 (3.7623)  weight_decay: 0.0500 (0.0500)  time: 0.2929  data: 0.0007  max mem: 21002
Epoch: [210]  [ 50/312]  eta: 0:03:07  lr: 0.000934  min_lr: 0.000934  loss: 4.0674 (3.7958)  weight_decay: 0.0500 (0.0500)  time: 0.5078  data: 0.0703  max mem: 21002
Epoch: [210]  [ 60/312]  eta: 0:02:42  lr: 0.000933  min_lr: 0.000933  loss: 3.9818 (3.7615)  weight_decay: 0.0500 (0.0500)  time: 0.5076  data: 0.0746  max mem: 21002
Epoch: [210]  [ 70/312]  eta: 0:02:38  lr: 0.000932  min_lr: 0.000932  loss: 3.8258 (3.7956)  weight_decay: 0.0500 (0.0500)  time: 0.5011  data: 0.0663  max mem: 21002
Epoch: [210]  [ 80/312]  eta: 0:02:23  lr: 0.000932  min_lr: 0.000932  loss: 3.8470 (3.7879)  weight_decay: 0.0500 (0.0500)  time: 0.5348  data: 0.0645  max mem: 21002
Epoch: [210]  [ 90/312]  eta: 0:02:16  lr: 0.000931  min_lr: 0.000931  loss: 3.8656 (3.7829)  weight_decay: 0.0500 (0.0500)  time: 0.4659  data: 0.0567  max mem: 21002
Epoch: [210]  [100/312]  eta: 0:02:11  lr: 0.000931  min_lr: 0.000931  loss: 3.9927 (3.7935)  weight_decay: 0.0500 (0.0500)  time: 0.6239  data: 0.1135  max mem: 21002
Epoch: [210]  [110/312]  eta: 0:01:59  lr: 0.000930  min_lr: 0.000930  loss: 3.8976 (3.8045)  weight_decay: 0.0500 (0.0500)  time: 0.4996  data: 0.0789  max mem: 21002
Epoch: [210]  [120/312]  eta: 0:01:54  lr: 0.000929  min_lr: 0.000929  loss: 3.6740 (3.7813)  weight_decay: 0.0500 (0.0500)  time: 0.4876  data: 0.0700  max mem: 21002
Epoch: [210]  [130/312]  eta: 0:01:49  lr: 0.000929  min_lr: 0.000929  loss: 3.6080 (3.7782)  weight_decay: 0.0500 (0.0500)  time: 0.6625  data: 0.1085  max mem: 21002
Epoch: [210]  [140/312]  eta: 0:01:41  lr: 0.000928  min_lr: 0.000928  loss: 3.9429 (3.7885)  weight_decay: 0.0500 (0.0500)  time: 0.5333  data: 0.1143  max mem: 21002
Epoch: [210]  [150/312]  eta: 0:01:35  lr: 0.000928  min_lr: 0.000928  loss: 3.9996 (3.7941)  weight_decay: 0.0500 (0.0500)  time: 0.5069  data: 0.0809  max mem: 21002
Epoch: [210]  [160/312]  eta: 0:01:28  lr: 0.000927  min_lr: 0.000927  loss: 3.9051 (3.7864)  weight_decay: 0.0500 (0.0500)  time: 0.5294  data: 0.0416  max mem: 21002
Epoch: [210]  [170/312]  eta: 0:01:22  lr: 0.000926  min_lr: 0.000926  loss: 3.5929 (3.7632)  weight_decay: 0.0500 (0.0500)  time: 0.4892  data: 0.0465  max mem: 21002
Epoch: [210]  [180/312]  eta: 0:01:16  lr: 0.000926  min_lr: 0.000926  loss: 3.4695 (3.7607)  weight_decay: 0.0500 (0.0500)  time: 0.5798  data: 0.0581  max mem: 21002
Epoch: [210]  [190/312]  eta: 0:01:09  lr: 0.000925  min_lr: 0.000925  loss: 3.5997 (3.7532)  weight_decay: 0.0500 (0.0500)  time: 0.4734  data: 0.0473  max mem: 21002
Epoch: [210]  [200/312]  eta: 0:01:04  lr: 0.000925  min_lr: 0.000925  loss: 3.5997 (3.7508)  weight_decay: 0.0500 (0.0500)  time: 0.5055  data: 0.0457  max mem: 21002
Epoch: [210]  [210/312]  eta: 0:00:58  lr: 0.000924  min_lr: 0.000924  loss: 3.5820 (3.7465)  weight_decay: 0.0500 (0.0500)  time: 0.6424  data: 0.0612  max mem: 21002
Epoch: [210]  [220/312]  eta: 0:00:52  lr: 0.000923  min_lr: 0.000923  loss: 3.7291 (3.7490)  weight_decay: 0.0500 (0.0500)  time: 0.4831  data: 0.0534  max mem: 21002
Epoch: [210]  [230/312]  eta: 0:00:46  lr: 0.000923  min_lr: 0.000923  loss: 3.5839 (3.7394)  weight_decay: 0.0500 (0.0500)  time: 0.5121  data: 0.0505  max mem: 21002
Epoch: [210]  [240/312]  eta: 0:00:40  lr: 0.000922  min_lr: 0.000922  loss: 3.8611 (3.7427)  weight_decay: 0.0500 (0.0500)  time: 0.5719  data: 0.0697  max mem: 21002
Epoch: [210]  [250/312]  eta: 0:00:35  lr: 0.000922  min_lr: 0.000922  loss: 3.9419 (3.7310)  weight_decay: 0.0500 (0.0500)  time: 0.5199  data: 0.0799  max mem: 21002
Epoch: [210]  [260/312]  eta: 0:00:29  lr: 0.000921  min_lr: 0.000921  loss: 3.8529 (3.7298)  weight_decay: 0.0500 (0.0500)  time: 0.6129  data: 0.0705  max mem: 21002
Epoch: [210]  [270/312]  eta: 0:00:23  lr: 0.000920  min_lr: 0.000920  loss: 4.0470 (3.7404)  weight_decay: 0.0500 (0.0500)  time: 0.5040  data: 0.0534  max mem: 21002
Epoch: [210]  [280/312]  eta: 0:00:18  lr: 0.000920  min_lr: 0.000920  loss: 4.1349 (3.7437)  weight_decay: 0.0500 (0.0500)  time: 0.5350  data: 0.0664  max mem: 21002
Epoch: [210]  [290/312]  eta: 0:00:12  lr: 0.000919  min_lr: 0.000919  loss: 4.0270 (3.7518)  weight_decay: 0.0500 (0.0500)  time: 0.6395  data: 0.0475  max mem: 21002
Epoch: [210]  [300/312]  eta: 0:00:06  lr: 0.000918  min_lr: 0.000918  loss: 3.9139 (3.7515)  weight_decay: 0.0500 (0.0500)  time: 0.4302  data: 0.0241  max mem: 21002
Epoch: [210]  [310/312]  eta: 0:00:01  lr: 0.000918  min_lr: 0.000918  loss: 3.7510 (3.7470)  weight_decay: 0.0500 (0.0500)  time: 0.3000  data: 0.0188  max mem: 21002
Epoch: [210]  [311/312]  eta: 0:00:00  lr: 0.000918  min_lr: 0.000918  loss: 3.7342 (3.7470)  weight_decay: 0.0500 (0.0500)  time: 0.2819  data: 0.0001  max mem: 21002
Epoch: [210] Total time: 0:02:52 (0.5522 s / it)
Averaged stats: lr: 0.000918  min_lr: 0.000918  loss: 3.7342 (3.7707)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.4654 (1.4654)  acc1: 74.6094 (74.6094)  acc5: 92.0573 (92.0573)  time: 8.6858  data: 8.5671  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.7829 (1.6661)  acc1: 66.2760 (66.7680)  acc5: 86.7188 (87.7120)  time: 1.0550  data: 0.9574  max mem: 21002
Test: Total time: 0:00:09 (1.0830 s / it)
* Acc@1 66.274 Acc@5 87.874 loss 1.673
Accuracy of the model on the 50000 test images: 66.3%
Max accuracy: 66.79%
Epoch: [211]  [  0/312]  eta: 0:59:53  lr: 0.000918  min_lr: 0.000918  loss: 4.0710 (4.0710)  weight_decay: 0.0500 (0.0500)  time: 11.5167  data: 9.4570  max mem: 21002
Epoch: [211]  [ 10/312]  eta: 0:08:29  lr: 0.000917  min_lr: 0.000917  loss: 3.9208 (3.7764)  weight_decay: 0.0500 (0.0500)  time: 1.6877  data: 1.0261  max mem: 21002
Epoch: [211]  [ 20/312]  eta: 0:05:04  lr: 0.000917  min_lr: 0.000917  loss: 3.7828 (3.7612)  weight_decay: 0.0500 (0.0500)  time: 0.5193  data: 0.0919  max mem: 21002
Epoch: [211]  [ 30/312]  eta: 0:03:45  lr: 0.000916  min_lr: 0.000916  loss: 3.6855 (3.7591)  weight_decay: 0.0500 (0.0500)  time: 0.3104  data: 0.0008  max mem: 21002
Epoch: [211]  [ 40/312]  eta: 0:03:08  lr: 0.000915  min_lr: 0.000915  loss: 3.7302 (3.7343)  weight_decay: 0.0500 (0.0500)  time: 0.3225  data: 0.0008  max mem: 21002
Epoch: [211]  [ 50/312]  eta: 0:02:57  lr: 0.000915  min_lr: 0.000915  loss: 3.5814 (3.7065)  weight_decay: 0.0500 (0.0500)  time: 0.4855  data: 0.0946  max mem: 21002
Epoch: [211]  [ 60/312]  eta: 0:02:35  lr: 0.000914  min_lr: 0.000914  loss: 3.8240 (3.7235)  weight_decay: 0.0500 (0.0500)  time: 0.4688  data: 0.1119  max mem: 21002
Epoch: [211]  [ 70/312]  eta: 0:02:31  lr: 0.000914  min_lr: 0.000914  loss: 3.7722 (3.7206)  weight_decay: 0.0500 (0.0500)  time: 0.4941  data: 0.1331  max mem: 21002
Epoch: [211]  [ 80/312]  eta: 0:02:25  lr: 0.000913  min_lr: 0.000913  loss: 3.9608 (3.7290)  weight_decay: 0.0500 (0.0500)  time: 0.6491  data: 0.1861  max mem: 21002
Epoch: [211]  [ 90/312]  eta: 0:02:11  lr: 0.000912  min_lr: 0.000912  loss: 3.8460 (3.7296)  weight_decay: 0.0500 (0.0500)  time: 0.4848  data: 0.0909  max mem: 21002
Epoch: [211]  [100/312]  eta: 0:02:07  lr: 0.000912  min_lr: 0.000912  loss: 3.7772 (3.7309)  weight_decay: 0.0500 (0.0500)  time: 0.4925  data: 0.1000  max mem: 21002
Epoch: [211]  [110/312]  eta: 0:01:57  lr: 0.000911  min_lr: 0.000911  loss: 3.8182 (3.7319)  weight_decay: 0.0500 (0.0500)  time: 0.5131  data: 0.1114  max mem: 21002
Epoch: [211]  [120/312]  eta: 0:01:52  lr: 0.000911  min_lr: 0.000911  loss: 3.4100 (3.7059)  weight_decay: 0.0500 (0.0500)  time: 0.5097  data: 0.1009  max mem: 21002
Epoch: [211]  [130/312]  eta: 0:01:47  lr: 0.000910  min_lr: 0.000910  loss: 3.3585 (3.7035)  weight_decay: 0.0500 (0.0500)  time: 0.6508  data: 0.1847  max mem: 21002
Epoch: [211]  [140/312]  eta: 0:01:39  lr: 0.000909  min_lr: 0.000909  loss: 3.8320 (3.7131)  weight_decay: 0.0500 (0.0500)  time: 0.5230  data: 0.1157  max mem: 21002
Epoch: [211]  [150/312]  eta: 0:01:34  lr: 0.000909  min_lr: 0.000909  loss: 3.9531 (3.7255)  weight_decay: 0.0500 (0.0500)  time: 0.5481  data: 0.0560  max mem: 21002
Epoch: [211]  [160/312]  eta: 0:01:28  lr: 0.000908  min_lr: 0.000908  loss: 4.0654 (3.7442)  weight_decay: 0.0500 (0.0500)  time: 0.6380  data: 0.1049  max mem: 21002
Epoch: [211]  [170/312]  eta: 0:01:21  lr: 0.000908  min_lr: 0.000908  loss: 3.8987 (3.7461)  weight_decay: 0.0500 (0.0500)  time: 0.4657  data: 0.0497  max mem: 21002
Epoch: [211]  [180/312]  eta: 0:01:15  lr: 0.000907  min_lr: 0.000907  loss: 3.8362 (3.7559)  weight_decay: 0.0500 (0.0500)  time: 0.4929  data: 0.0610  max mem: 21002
Epoch: [211]  [190/312]  eta: 0:01:08  lr: 0.000906  min_lr: 0.000906  loss: 3.6810 (3.7471)  weight_decay: 0.0500 (0.0500)  time: 0.5159  data: 0.0717  max mem: 21002
Epoch: [211]  [200/312]  eta: 0:01:04  lr: 0.000906  min_lr: 0.000906  loss: 3.5517 (3.7423)  weight_decay: 0.0500 (0.0500)  time: 0.5549  data: 0.0463  max mem: 21002
Epoch: [211]  [210/312]  eta: 0:00:58  lr: 0.000905  min_lr: 0.000905  loss: 3.9115 (3.7449)  weight_decay: 0.0500 (0.0500)  time: 0.6298  data: 0.0799  max mem: 21002
Epoch: [211]  [220/312]  eta: 0:00:52  lr: 0.000905  min_lr: 0.000905  loss: 3.9115 (3.7409)  weight_decay: 0.0500 (0.0500)  time: 0.5184  data: 0.0869  max mem: 21002
Epoch: [211]  [230/312]  eta: 0:00:46  lr: 0.000904  min_lr: 0.000904  loss: 3.8833 (3.7452)  weight_decay: 0.0500 (0.0500)  time: 0.5449  data: 0.0485  max mem: 21002
Epoch: [211]  [240/312]  eta: 0:00:40  lr: 0.000903  min_lr: 0.000903  loss: 3.8243 (3.7386)  weight_decay: 0.0500 (0.0500)  time: 0.5884  data: 0.0814  max mem: 21002
Epoch: [211]  [250/312]  eta: 0:00:34  lr: 0.000903  min_lr: 0.000903  loss: 3.9176 (3.7511)  weight_decay: 0.0500 (0.0500)  time: 0.4934  data: 0.0965  max mem: 21002
Epoch: [211]  [260/312]  eta: 0:00:29  lr: 0.000902  min_lr: 0.000902  loss: 4.0691 (3.7478)  weight_decay: 0.0500 (0.0500)  time: 0.5012  data: 0.0825  max mem: 21002
Epoch: [211]  [270/312]  eta: 0:00:23  lr: 0.000902  min_lr: 0.000902  loss: 3.8756 (3.7514)  weight_decay: 0.0500 (0.0500)  time: 0.4970  data: 0.0819  max mem: 21002
Epoch: [211]  [280/312]  eta: 0:00:17  lr: 0.000901  min_lr: 0.000901  loss: 3.8756 (3.7580)  weight_decay: 0.0500 (0.0500)  time: 0.5227  data: 0.0723  max mem: 21002
Epoch: [211]  [290/312]  eta: 0:00:12  lr: 0.000900  min_lr: 0.000900  loss: 3.8273 (3.7594)  weight_decay: 0.0500 (0.0500)  time: 0.5727  data: 0.0975  max mem: 21002
Epoch: [211]  [300/312]  eta: 0:00:06  lr: 0.000900  min_lr: 0.000900  loss: 3.8107 (3.7613)  weight_decay: 0.0500 (0.0500)  time: 0.4273  data: 0.0657  max mem: 21002
Epoch: [211]  [310/312]  eta: 0:00:01  lr: 0.000899  min_lr: 0.000899  loss: 3.7749 (3.7532)  weight_decay: 0.0500 (0.0500)  time: 0.3157  data: 0.0197  max mem: 21002
Epoch: [211]  [311/312]  eta: 0:00:00  lr: 0.000899  min_lr: 0.000899  loss: 3.7749 (3.7545)  weight_decay: 0.0500 (0.0500)  time: 0.3154  data: 0.0197  max mem: 21002
Epoch: [211] Total time: 0:02:50 (0.5463 s / it)
Averaged stats: lr: 0.000899  min_lr: 0.000899  loss: 3.7749 (3.7534)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 1.5193 (1.5193)  acc1: 72.5260 (72.5260)  acc5: 91.2760 (91.2760)  time: 8.9941  data: 8.8754  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.7705 (1.6917)  acc1: 65.8854 (66.2720)  acc5: 87.6302 (87.9360)  time: 1.0841  data: 0.9862  max mem: 21002
Test: Total time: 0:00:09 (1.1060 s / it)
* Acc@1 66.224 Acc@5 88.020 loss 1.703
Accuracy of the model on the 50000 test images: 66.2%
Max accuracy: 66.79%
Epoch: [212]  [  0/312]  eta: 1:02:59  lr: 0.000899  min_lr: 0.000899  loss: 3.6665 (3.6665)  weight_decay: 0.0500 (0.0500)  time: 12.1146  data: 9.1492  max mem: 21002
Epoch: [212]  [ 10/312]  eta: 0:08:22  lr: 0.000898  min_lr: 0.000898  loss: 3.6665 (3.5729)  weight_decay: 0.0500 (0.0500)  time: 1.6647  data: 0.9689  max mem: 21002
Epoch: [212]  [ 20/312]  eta: 0:05:13  lr: 0.000898  min_lr: 0.000898  loss: 3.9667 (3.7769)  weight_decay: 0.0500 (0.0500)  time: 0.5224  data: 0.1272  max mem: 21002
Epoch: [212]  [ 30/312]  eta: 0:03:51  lr: 0.000897  min_lr: 0.000897  loss: 4.0377 (3.8279)  weight_decay: 0.0500 (0.0500)  time: 0.3585  data: 0.0520  max mem: 21002
Epoch: [212]  [ 40/312]  eta: 0:03:12  lr: 0.000897  min_lr: 0.000897  loss: 4.0360 (3.8307)  weight_decay: 0.0500 (0.0500)  time: 0.3191  data: 0.0015  max mem: 21002
Epoch: [212]  [ 50/312]  eta: 0:03:02  lr: 0.000896  min_lr: 0.000896  loss: 3.6150 (3.7763)  weight_decay: 0.0500 (0.0500)  time: 0.5010  data: 0.0901  max mem: 21002
Epoch: [212]  [ 60/312]  eta: 0:02:39  lr: 0.000895  min_lr: 0.000895  loss: 3.4830 (3.7423)  weight_decay: 0.0500 (0.0500)  time: 0.4855  data: 0.1036  max mem: 21002
Epoch: [212]  [ 70/312]  eta: 0:02:33  lr: 0.000895  min_lr: 0.000895  loss: 3.7115 (3.7781)  weight_decay: 0.0500 (0.0500)  time: 0.4825  data: 0.0765  max mem: 21002
Epoch: [212]  [ 80/312]  eta: 0:02:25  lr: 0.000894  min_lr: 0.000894  loss: 3.9926 (3.7878)  weight_decay: 0.0500 (0.0500)  time: 0.6157  data: 0.1327  max mem: 21002
Epoch: [212]  [ 90/312]  eta: 0:02:14  lr: 0.000894  min_lr: 0.000894  loss: 3.9358 (3.7936)  weight_decay: 0.0500 (0.0500)  time: 0.5072  data: 0.0845  max mem: 21002
Epoch: [212]  [100/312]  eta: 0:02:09  lr: 0.000893  min_lr: 0.000893  loss: 3.9041 (3.7877)  weight_decay: 0.0500 (0.0500)  time: 0.5408  data: 0.0747  max mem: 21002
Epoch: [212]  [110/312]  eta: 0:01:57  lr: 0.000892  min_lr: 0.000892  loss: 3.9396 (3.7937)  weight_decay: 0.0500 (0.0500)  time: 0.4748  data: 0.0703  max mem: 21002
Epoch: [212]  [120/312]  eta: 0:01:53  lr: 0.000892  min_lr: 0.000892  loss: 3.9396 (3.8029)  weight_decay: 0.0500 (0.0500)  time: 0.5005  data: 0.0701  max mem: 21002
Epoch: [212]  [130/312]  eta: 0:01:49  lr: 0.000891  min_lr: 0.000891  loss: 4.0072 (3.8167)  weight_decay: 0.0500 (0.0500)  time: 0.6900  data: 0.1109  max mem: 21002
Epoch: [212]  [140/312]  eta: 0:01:39  lr: 0.000891  min_lr: 0.000891  loss: 4.0287 (3.8150)  weight_decay: 0.0500 (0.0500)  time: 0.4823  data: 0.0503  max mem: 21002
Epoch: [212]  [150/312]  eta: 0:01:35  lr: 0.000890  min_lr: 0.000890  loss: 3.8581 (3.8121)  weight_decay: 0.0500 (0.0500)  time: 0.5019  data: 0.0682  max mem: 21002
Epoch: [212]  [160/312]  eta: 0:01:28  lr: 0.000889  min_lr: 0.000889  loss: 3.8117 (3.7921)  weight_decay: 0.0500 (0.0500)  time: 0.6146  data: 0.1116  max mem: 21002
Epoch: [212]  [170/312]  eta: 0:01:21  lr: 0.000889  min_lr: 0.000889  loss: 3.9758 (3.8090)  weight_decay: 0.0500 (0.0500)  time: 0.4700  data: 0.0442  max mem: 21002
Epoch: [212]  [180/312]  eta: 0:01:16  lr: 0.000888  min_lr: 0.000888  loss: 4.0641 (3.8154)  weight_decay: 0.0500 (0.0500)  time: 0.5743  data: 0.0531  max mem: 21002
Epoch: [212]  [190/312]  eta: 0:01:09  lr: 0.000888  min_lr: 0.000888  loss: 3.9891 (3.8117)  weight_decay: 0.0500 (0.0500)  time: 0.5067  data: 0.0591  max mem: 21002
Epoch: [212]  [200/312]  eta: 0:01:04  lr: 0.000887  min_lr: 0.000887  loss: 3.8884 (3.8198)  weight_decay: 0.0500 (0.0500)  time: 0.4848  data: 0.0438  max mem: 21002
Epoch: [212]  [210/312]  eta: 0:00:59  lr: 0.000886  min_lr: 0.000886  loss: 3.8884 (3.8115)  weight_decay: 0.0500 (0.0500)  time: 0.7093  data: 0.0905  max mem: 21002
Epoch: [212]  [220/312]  eta: 0:00:52  lr: 0.000886  min_lr: 0.000886  loss: 3.8952 (3.8168)  weight_decay: 0.0500 (0.0500)  time: 0.5171  data: 0.0534  max mem: 21002
Epoch: [212]  [230/312]  eta: 0:00:46  lr: 0.000885  min_lr: 0.000885  loss: 3.8914 (3.8111)  weight_decay: 0.0500 (0.0500)  time: 0.4873  data: 0.0225  max mem: 21002
Epoch: [212]  [240/312]  eta: 0:00:40  lr: 0.000885  min_lr: 0.000885  loss: 3.5735 (3.7951)  weight_decay: 0.0500 (0.0500)  time: 0.5921  data: 0.0425  max mem: 21002
Epoch: [212]  [250/312]  eta: 0:00:35  lr: 0.000884  min_lr: 0.000884  loss: 3.6702 (3.7965)  weight_decay: 0.0500 (0.0500)  time: 0.4876  data: 0.0207  max mem: 21002
Epoch: [212]  [260/312]  eta: 0:00:29  lr: 0.000883  min_lr: 0.000883  loss: 3.6406 (3.7804)  weight_decay: 0.0500 (0.0500)  time: 0.6058  data: 0.0132  max mem: 21002
Epoch: [212]  [270/312]  eta: 0:00:23  lr: 0.000883  min_lr: 0.000883  loss: 3.5384 (3.7775)  weight_decay: 0.0500 (0.0500)  time: 0.5146  data: 0.0132  max mem: 21002
Epoch: [212]  [280/312]  eta: 0:00:18  lr: 0.000882  min_lr: 0.000882  loss: 3.9472 (3.7831)  weight_decay: 0.0500 (0.0500)  time: 0.4878  data: 0.0646  max mem: 21002
Epoch: [212]  [290/312]  eta: 0:00:12  lr: 0.000882  min_lr: 0.000882  loss: 3.7962 (3.7769)  weight_decay: 0.0500 (0.0500)  time: 0.6332  data: 0.0644  max mem: 21002
Epoch: [212]  [300/312]  eta: 0:00:06  lr: 0.000881  min_lr: 0.000881  loss: 3.7009 (3.7813)  weight_decay: 0.0500 (0.0500)  time: 0.4305  data: 0.0002  max mem: 21002
Epoch: [212]  [310/312]  eta: 0:00:01  lr: 0.000880  min_lr: 0.000880  loss: 3.9568 (3.7799)  weight_decay: 0.0500 (0.0500)  time: 0.2772  data: 0.0001  max mem: 21002
Epoch: [212]  [311/312]  eta: 0:00:00  lr: 0.000880  min_lr: 0.000880  loss: 3.9568 (3.7810)  weight_decay: 0.0500 (0.0500)  time: 0.2772  data: 0.0001  max mem: 21002
Epoch: [212] Total time: 0:02:51 (0.5506 s / it)
Averaged stats: lr: 0.000880  min_lr: 0.000880  loss: 3.9568 (3.7628)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.3871 (1.3871)  acc1: 75.7812 (75.7812)  acc5: 93.3594 (93.3594)  time: 8.5528  data: 8.4361  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.9009 (1.7149)  acc1: 64.0625 (66.5120)  acc5: 85.8073 (87.8240)  time: 1.0420  data: 0.9374  max mem: 21002
Test: Total time: 0:00:09 (1.0699 s / it)
* Acc@1 66.774 Acc@5 87.748 loss 1.723
Accuracy of the model on the 50000 test images: 66.8%
Max accuracy: 66.79%
Epoch: [213]  [  0/312]  eta: 1:03:00  lr: 0.000880  min_lr: 0.000880  loss: 4.5409 (4.5409)  weight_decay: 0.0500 (0.0500)  time: 12.1185  data: 9.6218  max mem: 21002
Epoch: [213]  [ 10/312]  eta: 0:08:36  lr: 0.000880  min_lr: 0.000880  loss: 3.9741 (3.9194)  weight_decay: 0.0500 (0.0500)  time: 1.7104  data: 0.9722  max mem: 21002
Epoch: [213]  [ 20/312]  eta: 0:05:08  lr: 0.000879  min_lr: 0.000879  loss: 3.9673 (3.9266)  weight_decay: 0.0500 (0.0500)  time: 0.5020  data: 0.0771  max mem: 21002
Epoch: [213]  [ 30/312]  eta: 0:03:48  lr: 0.000879  min_lr: 0.000879  loss: 3.8710 (3.8975)  weight_decay: 0.0500 (0.0500)  time: 0.3125  data: 0.0238  max mem: 21002
Epoch: [213]  [ 40/312]  eta: 0:03:06  lr: 0.000878  min_lr: 0.000878  loss: 3.7361 (3.8776)  weight_decay: 0.0500 (0.0500)  time: 0.2952  data: 0.0006  max mem: 21002
Epoch: [213]  [ 50/312]  eta: 0:03:00  lr: 0.000877  min_lr: 0.000877  loss: 3.8990 (3.8603)  weight_decay: 0.0500 (0.0500)  time: 0.4998  data: 0.1072  max mem: 21002
Epoch: [213]  [ 60/312]  eta: 0:02:37  lr: 0.000877  min_lr: 0.000877  loss: 3.9042 (3.8615)  weight_decay: 0.0500 (0.0500)  time: 0.5012  data: 0.1127  max mem: 21002
Epoch: [213]  [ 70/312]  eta: 0:02:28  lr: 0.000876  min_lr: 0.000876  loss: 3.7601 (3.8219)  weight_decay: 0.0500 (0.0500)  time: 0.4262  data: 0.1049  max mem: 21002
Epoch: [213]  [ 80/312]  eta: 0:02:23  lr: 0.000876  min_lr: 0.000876  loss: 3.6264 (3.8163)  weight_decay: 0.0500 (0.0500)  time: 0.6076  data: 0.1962  max mem: 21002
Epoch: [213]  [ 90/312]  eta: 0:02:10  lr: 0.000875  min_lr: 0.000875  loss: 3.7626 (3.8143)  weight_decay: 0.0500 (0.0500)  time: 0.5034  data: 0.0977  max mem: 21002
Epoch: [213]  [100/312]  eta: 0:02:06  lr: 0.000874  min_lr: 0.000874  loss: 3.8718 (3.8105)  weight_decay: 0.0500 (0.0500)  time: 0.5113  data: 0.1058  max mem: 21002
Epoch: [213]  [110/312]  eta: 0:01:55  lr: 0.000874  min_lr: 0.000874  loss: 3.8752 (3.8226)  weight_decay: 0.0500 (0.0500)  time: 0.5013  data: 0.1103  max mem: 21002
Epoch: [213]  [120/312]  eta: 0:01:52  lr: 0.000873  min_lr: 0.000873  loss: 3.8490 (3.8112)  weight_decay: 0.0500 (0.0500)  time: 0.5063  data: 0.0881  max mem: 21002
Epoch: [213]  [130/312]  eta: 0:01:47  lr: 0.000873  min_lr: 0.000873  loss: 3.7133 (3.8002)  weight_decay: 0.0500 (0.0500)  time: 0.6796  data: 0.1549  max mem: 21002
Epoch: [213]  [140/312]  eta: 0:01:37  lr: 0.000872  min_lr: 0.000872  loss: 3.9578 (3.7979)  weight_decay: 0.0500 (0.0500)  time: 0.4781  data: 0.0722  max mem: 21002
Epoch: [213]  [150/312]  eta: 0:01:33  lr: 0.000871  min_lr: 0.000871  loss: 3.9234 (3.7943)  weight_decay: 0.0500 (0.0500)  time: 0.4982  data: 0.0710  max mem: 21002
Epoch: [213]  [160/312]  eta: 0:01:27  lr: 0.000871  min_lr: 0.000871  loss: 3.8472 (3.7833)  weight_decay: 0.0500 (0.0500)  time: 0.6458  data: 0.0830  max mem: 21002
Epoch: [213]  [170/312]  eta: 0:01:20  lr: 0.000870  min_lr: 0.000870  loss: 3.6226 (3.7697)  weight_decay: 0.0500 (0.0500)  time: 0.4804  data: 0.0443  max mem: 21002
Epoch: [213]  [180/312]  eta: 0:01:15  lr: 0.000870  min_lr: 0.000870  loss: 3.8536 (3.7800)  weight_decay: 0.0500 (0.0500)  time: 0.5140  data: 0.0891  max mem: 21002
Epoch: [213]  [190/312]  eta: 0:01:08  lr: 0.000869  min_lr: 0.000869  loss: 3.9197 (3.7834)  weight_decay: 0.0500 (0.0500)  time: 0.5043  data: 0.0574  max mem: 21002
Epoch: [213]  [200/312]  eta: 0:01:03  lr: 0.000868  min_lr: 0.000868  loss: 3.9197 (3.7918)  weight_decay: 0.0500 (0.0500)  time: 0.5432  data: 0.0556  max mem: 21002
Epoch: [213]  [210/312]  eta: 0:00:58  lr: 0.000868  min_lr: 0.000868  loss: 3.6943 (3.7822)  weight_decay: 0.0500 (0.0500)  time: 0.6623  data: 0.0616  max mem: 21002
Epoch: [213]  [220/312]  eta: 0:00:51  lr: 0.000867  min_lr: 0.000867  loss: 3.5360 (3.7803)  weight_decay: 0.0500 (0.0500)  time: 0.4554  data: 0.0070  max mem: 21002
Epoch: [213]  [230/312]  eta: 0:00:46  lr: 0.000867  min_lr: 0.000867  loss: 4.0626 (3.7990)  weight_decay: 0.0500 (0.0500)  time: 0.4750  data: 0.0322  max mem: 21002
Epoch: [213]  [240/312]  eta: 0:00:40  lr: 0.000866  min_lr: 0.000866  loss: 4.0163 (3.7917)  weight_decay: 0.0500 (0.0500)  time: 0.6068  data: 0.0528  max mem: 21002
Epoch: [213]  [250/312]  eta: 0:00:34  lr: 0.000865  min_lr: 0.000865  loss: 3.8647 (3.7949)  weight_decay: 0.0500 (0.0500)  time: 0.5114  data: 0.0500  max mem: 21002
Epoch: [213]  [260/312]  eta: 0:00:29  lr: 0.000865  min_lr: 0.000865  loss: 3.8226 (3.7908)  weight_decay: 0.0500 (0.0500)  time: 0.5507  data: 0.0482  max mem: 21002
Epoch: [213]  [270/312]  eta: 0:00:23  lr: 0.000864  min_lr: 0.000864  loss: 3.6531 (3.7749)  weight_decay: 0.0500 (0.0500)  time: 0.4887  data: 0.0260  max mem: 21002
Epoch: [213]  [280/312]  eta: 0:00:17  lr: 0.000864  min_lr: 0.000864  loss: 3.6531 (3.7689)  weight_decay: 0.0500 (0.0500)  time: 0.5083  data: 0.0314  max mem: 21002
Epoch: [213]  [290/312]  eta: 0:00:12  lr: 0.000863  min_lr: 0.000863  loss: 3.9331 (3.7742)  weight_decay: 0.0500 (0.0500)  time: 0.6713  data: 0.0396  max mem: 21002
Epoch: [213]  [300/312]  eta: 0:00:06  lr: 0.000863  min_lr: 0.000863  loss: 3.9741 (3.7760)  weight_decay: 0.0500 (0.0500)  time: 0.4739  data: 0.0147  max mem: 21002
Epoch: [213]  [310/312]  eta: 0:00:01  lr: 0.000862  min_lr: 0.000862  loss: 3.5489 (3.7693)  weight_decay: 0.0500 (0.0500)  time: 0.2907  data: 0.0001  max mem: 21002
Epoch: [213]  [311/312]  eta: 0:00:00  lr: 0.000862  min_lr: 0.000862  loss: 3.5282 (3.7661)  weight_decay: 0.0500 (0.0500)  time: 0.2905  data: 0.0001  max mem: 21002
Epoch: [213] Total time: 0:02:50 (0.5461 s / it)
Averaged stats: lr: 0.000862  min_lr: 0.000862  loss: 3.5282 (3.7357)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.2928 (1.2928)  acc1: 75.5208 (75.5208)  acc5: 92.7083 (92.7083)  time: 8.4596  data: 8.3414  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.6940 (1.5702)  acc1: 66.0156 (67.4080)  acc5: 88.5417 (88.4000)  time: 1.0460  data: 0.9401  max mem: 21002
Test: Total time: 0:00:09 (1.0556 s / it)
* Acc@1 67.040 Acc@5 88.372 loss 1.584
Accuracy of the model on the 50000 test images: 67.0%
Max accuracy: 67.04%
Epoch: [214]  [  0/312]  eta: 1:00:30  lr: 0.000862  min_lr: 0.000862  loss: 2.9992 (2.9992)  weight_decay: 0.0500 (0.0500)  time: 11.6353  data: 10.5535  max mem: 21002
Epoch: [214]  [ 10/312]  eta: 0:08:13  lr: 0.000861  min_lr: 0.000861  loss: 3.9036 (3.6623)  weight_decay: 0.0500 (0.0500)  time: 1.6337  data: 1.0273  max mem: 21002
Epoch: [214]  [ 20/312]  eta: 0:04:53  lr: 0.000861  min_lr: 0.000861  loss: 3.9036 (3.6438)  weight_decay: 0.0500 (0.0500)  time: 0.4738  data: 0.0376  max mem: 21002
Epoch: [214]  [ 30/312]  eta: 0:03:38  lr: 0.000860  min_lr: 0.000860  loss: 3.8411 (3.6953)  weight_decay: 0.0500 (0.0500)  time: 0.3027  data: 0.0015  max mem: 21002
Epoch: [214]  [ 40/312]  eta: 0:03:09  lr: 0.000859  min_lr: 0.000859  loss: 3.8411 (3.7187)  weight_decay: 0.0500 (0.0500)  time: 0.3737  data: 0.0232  max mem: 21002
Epoch: [214]  [ 50/312]  eta: 0:03:01  lr: 0.000859  min_lr: 0.000859  loss: 3.6983 (3.7198)  weight_decay: 0.0500 (0.0500)  time: 0.5649  data: 0.1196  max mem: 21002
Epoch: [214]  [ 60/312]  eta: 0:02:39  lr: 0.000858  min_lr: 0.000858  loss: 3.8678 (3.7569)  weight_decay: 0.0500 (0.0500)  time: 0.4989  data: 0.1102  max mem: 21002
Epoch: [214]  [ 70/312]  eta: 0:02:32  lr: 0.000858  min_lr: 0.000858  loss: 3.7664 (3.7051)  weight_decay: 0.0500 (0.0500)  time: 0.4769  data: 0.1184  max mem: 21002
Epoch: [214]  [ 80/312]  eta: 0:02:25  lr: 0.000857  min_lr: 0.000857  loss: 3.6624 (3.7065)  weight_decay: 0.0500 (0.0500)  time: 0.6105  data: 0.1784  max mem: 21002
Epoch: [214]  [ 90/312]  eta: 0:02:13  lr: 0.000857  min_lr: 0.000857  loss: 3.7020 (3.6995)  weight_decay: 0.0500 (0.0500)  time: 0.4886  data: 0.1228  max mem: 21002
Epoch: [214]  [100/312]  eta: 0:02:10  lr: 0.000856  min_lr: 0.000856  loss: 3.7254 (3.6970)  weight_decay: 0.0500 (0.0500)  time: 0.5706  data: 0.1731  max mem: 21002
Epoch: [214]  [110/312]  eta: 0:01:58  lr: 0.000855  min_lr: 0.000855  loss: 3.5735 (3.6833)  weight_decay: 0.0500 (0.0500)  time: 0.5226  data: 0.1233  max mem: 21002
Epoch: [214]  [120/312]  eta: 0:01:55  lr: 0.000855  min_lr: 0.000855  loss: 3.6916 (3.6969)  weight_decay: 0.0500 (0.0500)  time: 0.5178  data: 0.0899  max mem: 21002
Epoch: [214]  [130/312]  eta: 0:01:48  lr: 0.000854  min_lr: 0.000854  loss: 3.7806 (3.6808)  weight_decay: 0.0500 (0.0500)  time: 0.6456  data: 0.1207  max mem: 21002
Epoch: [214]  [140/312]  eta: 0:01:39  lr: 0.000854  min_lr: 0.000854  loss: 3.6883 (3.6824)  weight_decay: 0.0500 (0.0500)  time: 0.4435  data: 0.0594  max mem: 21002
Epoch: [214]  [150/312]  eta: 0:01:34  lr: 0.000853  min_lr: 0.000853  loss: 3.6883 (3.6894)  weight_decay: 0.0500 (0.0500)  time: 0.5181  data: 0.1273  max mem: 21002
Epoch: [214]  [160/312]  eta: 0:01:27  lr: 0.000852  min_lr: 0.000852  loss: 3.8261 (3.6879)  weight_decay: 0.0500 (0.0500)  time: 0.5866  data: 0.1058  max mem: 21002
Epoch: [214]  [170/312]  eta: 0:01:21  lr: 0.000852  min_lr: 0.000852  loss: 4.0563 (3.7061)  weight_decay: 0.0500 (0.0500)  time: 0.4821  data: 0.0983  max mem: 21002
Epoch: [214]  [180/312]  eta: 0:01:16  lr: 0.000851  min_lr: 0.000851  loss: 4.0341 (3.7087)  weight_decay: 0.0500 (0.0500)  time: 0.5737  data: 0.1778  max mem: 21002
Epoch: [214]  [190/312]  eta: 0:01:08  lr: 0.000851  min_lr: 0.000851  loss: 3.6440 (3.7086)  weight_decay: 0.0500 (0.0500)  time: 0.4749  data: 0.0864  max mem: 21002
Epoch: [214]  [200/312]  eta: 0:01:03  lr: 0.000850  min_lr: 0.000850  loss: 3.6440 (3.6986)  weight_decay: 0.0500 (0.0500)  time: 0.4820  data: 0.0954  max mem: 21002
Epoch: [214]  [210/312]  eta: 0:00:58  lr: 0.000849  min_lr: 0.000849  loss: 3.7233 (3.7016)  weight_decay: 0.0500 (0.0500)  time: 0.6510  data: 0.1258  max mem: 21002
Epoch: [214]  [220/312]  eta: 0:00:51  lr: 0.000849  min_lr: 0.000849  loss: 3.9225 (3.7119)  weight_decay: 0.0500 (0.0500)  time: 0.5101  data: 0.0863  max mem: 21002
Epoch: [214]  [230/312]  eta: 0:00:46  lr: 0.000848  min_lr: 0.000848  loss: 4.0852 (3.7216)  weight_decay: 0.0500 (0.0500)  time: 0.5374  data: 0.1305  max mem: 21002
Epoch: [214]  [240/312]  eta: 0:00:40  lr: 0.000848  min_lr: 0.000848  loss: 3.8930 (3.7210)  weight_decay: 0.0500 (0.0500)  time: 0.5778  data: 0.0752  max mem: 21002
Epoch: [214]  [250/312]  eta: 0:00:34  lr: 0.000847  min_lr: 0.000847  loss: 3.7422 (3.7140)  weight_decay: 0.0500 (0.0500)  time: 0.4660  data: 0.0644  max mem: 21002
Epoch: [214]  [260/312]  eta: 0:00:29  lr: 0.000846  min_lr: 0.000846  loss: 3.9271 (3.7239)  weight_decay: 0.0500 (0.0500)  time: 0.5785  data: 0.1580  max mem: 21002
Epoch: [214]  [270/312]  eta: 0:00:23  lr: 0.000846  min_lr: 0.000846  loss: 3.8985 (3.7174)  weight_decay: 0.0500 (0.0500)  time: 0.5090  data: 0.0944  max mem: 21002
Epoch: [214]  [280/312]  eta: 0:00:17  lr: 0.000845  min_lr: 0.000845  loss: 3.6410 (3.7115)  weight_decay: 0.0500 (0.0500)  time: 0.4496  data: 0.0602  max mem: 21002
Epoch: [214]  [290/312]  eta: 0:00:12  lr: 0.000845  min_lr: 0.000845  loss: 3.6973 (3.7069)  weight_decay: 0.0500 (0.0500)  time: 0.6439  data: 0.1325  max mem: 21002
Epoch: [214]  [300/312]  eta: 0:00:06  lr: 0.000844  min_lr: 0.000844  loss: 3.4967 (3.7014)  weight_decay: 0.0500 (0.0500)  time: 0.4922  data: 0.0728  max mem: 21002
Epoch: [214]  [310/312]  eta: 0:00:01  lr: 0.000844  min_lr: 0.000844  loss: 3.7799 (3.7066)  weight_decay: 0.0500 (0.0500)  time: 0.2773  data: 0.0001  max mem: 21002
Epoch: [214]  [311/312]  eta: 0:00:00  lr: 0.000844  min_lr: 0.000844  loss: 3.7799 (3.7078)  weight_decay: 0.0500 (0.0500)  time: 0.2772  data: 0.0001  max mem: 21002
Epoch: [214] Total time: 0:02:50 (0.5469 s / it)
Averaged stats: lr: 0.000844  min_lr: 0.000844  loss: 3.7799 (3.7328)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.3327 (1.3327)  acc1: 73.3073 (73.3073)  acc5: 91.2760 (91.2760)  time: 8.6094  data: 8.4906  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.8879 (1.7096)  acc1: 62.3698 (64.7840)  acc5: 86.7188 (86.6720)  time: 1.0496  data: 0.9516  max mem: 21002
Test: Total time: 0:00:09 (1.0799 s / it)
* Acc@1 64.976 Acc@5 87.044 loss 1.715
Accuracy of the model on the 50000 test images: 65.0%
Max accuracy: 67.04%
Epoch: [215]  [  0/312]  eta: 1:02:03  lr: 0.000843  min_lr: 0.000843  loss: 4.4022 (4.4022)  weight_decay: 0.0500 (0.0500)  time: 11.9354  data: 8.7780  max mem: 21002
Epoch: [215]  [ 10/312]  eta: 0:08:03  lr: 0.000843  min_lr: 0.000843  loss: 4.0177 (3.9700)  weight_decay: 0.0500 (0.0500)  time: 1.6025  data: 0.9886  max mem: 21002
Epoch: [215]  [ 20/312]  eta: 0:05:20  lr: 0.000842  min_lr: 0.000842  loss: 3.8800 (3.8295)  weight_decay: 0.0500 (0.0500)  time: 0.5573  data: 0.1910  max mem: 21002
Epoch: [215]  [ 30/312]  eta: 0:03:56  lr: 0.000842  min_lr: 0.000842  loss: 3.8172 (3.8307)  weight_decay: 0.0500 (0.0500)  time: 0.4184  data: 0.0866  max mem: 21002
Epoch: [215]  [ 40/312]  eta: 0:03:19  lr: 0.000841  min_lr: 0.000841  loss: 3.8828 (3.8092)  weight_decay: 0.0500 (0.0500)  time: 0.3491  data: 0.0184  max mem: 21002
Epoch: [215]  [ 50/312]  eta: 0:03:08  lr: 0.000841  min_lr: 0.000841  loss: 4.0199 (3.8334)  weight_decay: 0.0500 (0.0500)  time: 0.5334  data: 0.1238  max mem: 21002
Epoch: [215]  [ 60/312]  eta: 0:02:43  lr: 0.000840  min_lr: 0.000840  loss: 4.1039 (3.8592)  weight_decay: 0.0500 (0.0500)  time: 0.4784  data: 0.1108  max mem: 21002
Epoch: [215]  [ 70/312]  eta: 0:02:39  lr: 0.000839  min_lr: 0.000839  loss: 4.0270 (3.8792)  weight_decay: 0.0500 (0.0500)  time: 0.5090  data: 0.1344  max mem: 21002
Epoch: [215]  [ 80/312]  eta: 0:02:28  lr: 0.000839  min_lr: 0.000839  loss: 4.0085 (3.8619)  weight_decay: 0.0500 (0.0500)  time: 0.6033  data: 0.1654  max mem: 21002
Epoch: [215]  [ 90/312]  eta: 0:02:17  lr: 0.000838  min_lr: 0.000838  loss: 3.8184 (3.8569)  weight_decay: 0.0500 (0.0500)  time: 0.4670  data: 0.1073  max mem: 21002
Epoch: [215]  [100/312]  eta: 0:02:12  lr: 0.000838  min_lr: 0.000838  loss: 3.7623 (3.8273)  weight_decay: 0.0500 (0.0500)  time: 0.5682  data: 0.1924  max mem: 21002
Epoch: [215]  [110/312]  eta: 0:01:59  lr: 0.000837  min_lr: 0.000837  loss: 3.8500 (3.8414)  weight_decay: 0.0500 (0.0500)  time: 0.4863  data: 0.1214  max mem: 21002
Epoch: [215]  [120/312]  eta: 0:01:55  lr: 0.000836  min_lr: 0.000836  loss: 3.9840 (3.8500)  weight_decay: 0.0500 (0.0500)  time: 0.4931  data: 0.0924  max mem: 21002
Epoch: [215]  [130/312]  eta: 0:01:49  lr: 0.000836  min_lr: 0.000836  loss: 3.9246 (3.8485)  weight_decay: 0.0500 (0.0500)  time: 0.6373  data: 0.1614  max mem: 21002
Epoch: [215]  [140/312]  eta: 0:01:39  lr: 0.000835  min_lr: 0.000835  loss: 3.8338 (3.8469)  weight_decay: 0.0500 (0.0500)  time: 0.4350  data: 0.0748  max mem: 21002
Epoch: [215]  [150/312]  eta: 0:01:34  lr: 0.000835  min_lr: 0.000835  loss: 3.3277 (3.8014)  weight_decay: 0.0500 (0.0500)  time: 0.4909  data: 0.1165  max mem: 21002
Epoch: [215]  [160/312]  eta: 0:01:28  lr: 0.000834  min_lr: 0.000834  loss: 3.2207 (3.7899)  weight_decay: 0.0500 (0.0500)  time: 0.6062  data: 0.1803  max mem: 21002
Epoch: [215]  [170/312]  eta: 0:01:21  lr: 0.000833  min_lr: 0.000833  loss: 3.8201 (3.7921)  weight_decay: 0.0500 (0.0500)  time: 0.4839  data: 0.1354  max mem: 21002
Epoch: [215]  [180/312]  eta: 0:01:16  lr: 0.000833  min_lr: 0.000833  loss: 3.9717 (3.8041)  weight_decay: 0.0500 (0.0500)  time: 0.5660  data: 0.2079  max mem: 21002
Epoch: [215]  [190/312]  eta: 0:01:08  lr: 0.000832  min_lr: 0.000832  loss: 3.8115 (3.7953)  weight_decay: 0.0500 (0.0500)  time: 0.4900  data: 0.1421  max mem: 21002
Epoch: [215]  [200/312]  eta: 0:01:04  lr: 0.000832  min_lr: 0.000832  loss: 3.6953 (3.7953)  weight_decay: 0.0500 (0.0500)  time: 0.4967  data: 0.1419  max mem: 21002
Epoch: [215]  [210/312]  eta: 0:00:58  lr: 0.000831  min_lr: 0.000831  loss: 3.9382 (3.7948)  weight_decay: 0.0500 (0.0500)  time: 0.7026  data: 0.2510  max mem: 21002
Epoch: [215]  [220/312]  eta: 0:00:51  lr: 0.000831  min_lr: 0.000831  loss: 3.6870 (3.7896)  weight_decay: 0.0500 (0.0500)  time: 0.4929  data: 0.1097  max mem: 21002
Epoch: [215]  [230/312]  eta: 0:00:46  lr: 0.000830  min_lr: 0.000830  loss: 3.8061 (3.7881)  weight_decay: 0.0500 (0.0500)  time: 0.4562  data: 0.0890  max mem: 21002
Epoch: [215]  [240/312]  eta: 0:00:40  lr: 0.000829  min_lr: 0.000829  loss: 3.8502 (3.7816)  weight_decay: 0.0500 (0.0500)  time: 0.5814  data: 0.1593  max mem: 21002
Epoch: [215]  [250/312]  eta: 0:00:34  lr: 0.000829  min_lr: 0.000829  loss: 3.7038 (3.7811)  weight_decay: 0.0500 (0.0500)  time: 0.5039  data: 0.1218  max mem: 21002
Epoch: [215]  [260/312]  eta: 0:00:29  lr: 0.000828  min_lr: 0.000828  loss: 3.6518 (3.7708)  weight_decay: 0.0500 (0.0500)  time: 0.5633  data: 0.1570  max mem: 21002
Epoch: [215]  [270/312]  eta: 0:00:23  lr: 0.000828  min_lr: 0.000828  loss: 3.8151 (3.7784)  weight_decay: 0.0500 (0.0500)  time: 0.4714  data: 0.1063  max mem: 21002
Epoch: [215]  [280/312]  eta: 0:00:18  lr: 0.000827  min_lr: 0.000827  loss: 3.6238 (3.7636)  weight_decay: 0.0500 (0.0500)  time: 0.5162  data: 0.1196  max mem: 21002
Epoch: [215]  [290/312]  eta: 0:00:12  lr: 0.000827  min_lr: 0.000827  loss: 3.5888 (3.7647)  weight_decay: 0.0500 (0.0500)  time: 0.6474  data: 0.1939  max mem: 21002
Epoch: [215]  [300/312]  eta: 0:00:06  lr: 0.000826  min_lr: 0.000826  loss: 3.9188 (3.7684)  weight_decay: 0.0500 (0.0500)  time: 0.4140  data: 0.0746  max mem: 21002
Epoch: [215]  [310/312]  eta: 0:00:01  lr: 0.000825  min_lr: 0.000825  loss: 4.0217 (3.7759)  weight_decay: 0.0500 (0.0500)  time: 0.2773  data: 0.0001  max mem: 21002
Epoch: [215]  [311/312]  eta: 0:00:00  lr: 0.000825  min_lr: 0.000825  loss: 4.0217 (3.7771)  weight_decay: 0.0500 (0.0500)  time: 0.2772  data: 0.0001  max mem: 21002
Epoch: [215] Total time: 0:02:50 (0.5466 s / it)
Averaged stats: lr: 0.000825  min_lr: 0.000825  loss: 4.0217 (3.7392)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.1616 (1.1616)  acc1: 76.5625 (76.5625)  acc5: 93.2292 (93.2292)  time: 8.5033  data: 8.3853  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.7692 (1.6521)  acc1: 65.3646 (66.8640)  acc5: 87.8906 (88.0640)  time: 1.0298  data: 0.9318  max mem: 21002
Test: Total time: 0:00:09 (1.0395 s / it)
* Acc@1 66.772 Acc@5 88.034 loss 1.657
Accuracy of the model on the 50000 test images: 66.8%
Max accuracy: 67.04%
Epoch: [216]  [  0/312]  eta: 1:04:38  lr: 0.000825  min_lr: 0.000825  loss: 3.4630 (3.4630)  weight_decay: 0.0500 (0.0500)  time: 12.4325  data: 10.7914  max mem: 21002
Epoch: [216]  [ 10/312]  eta: 0:08:28  lr: 0.000825  min_lr: 0.000825  loss: 3.9654 (3.9361)  weight_decay: 0.0500 (0.0500)  time: 1.6825  data: 0.9955  max mem: 21002
Epoch: [216]  [ 20/312]  eta: 0:05:06  lr: 0.000824  min_lr: 0.000824  loss: 3.9172 (3.8655)  weight_decay: 0.0500 (0.0500)  time: 0.4810  data: 0.0407  max mem: 21002
Epoch: [216]  [ 30/312]  eta: 0:03:47  lr: 0.000823  min_lr: 0.000823  loss: 3.8876 (3.8935)  weight_decay: 0.0500 (0.0500)  time: 0.3237  data: 0.0330  max mem: 21002
Epoch: [216]  [ 40/312]  eta: 0:03:08  lr: 0.000823  min_lr: 0.000823  loss: 3.9175 (3.9092)  weight_decay: 0.0500 (0.0500)  time: 0.3163  data: 0.0229  max mem: 21002
Epoch: [216]  [ 50/312]  eta: 0:02:59  lr: 0.000822  min_lr: 0.000822  loss: 3.8893 (3.8304)  weight_decay: 0.0500 (0.0500)  time: 0.5012  data: 0.1936  max mem: 21002
Epoch: [216]  [ 60/312]  eta: 0:02:36  lr: 0.000822  min_lr: 0.000822  loss: 3.8266 (3.8425)  weight_decay: 0.0500 (0.0500)  time: 0.4750  data: 0.1722  max mem: 21002
Epoch: [216]  [ 70/312]  eta: 0:02:31  lr: 0.000821  min_lr: 0.000821  loss: 3.8796 (3.8201)  weight_decay: 0.0500 (0.0500)  time: 0.4790  data: 0.1743  max mem: 21002
Epoch: [216]  [ 80/312]  eta: 0:02:26  lr: 0.000821  min_lr: 0.000821  loss: 3.6541 (3.7916)  weight_decay: 0.0500 (0.0500)  time: 0.6718  data: 0.3284  max mem: 21002
Epoch: [216]  [ 90/312]  eta: 0:02:13  lr: 0.000820  min_lr: 0.000820  loss: 3.8050 (3.7895)  weight_decay: 0.0500 (0.0500)  time: 0.4972  data: 0.1731  max mem: 21002
Epoch: [216]  [100/312]  eta: 0:02:09  lr: 0.000819  min_lr: 0.000819  loss: 3.6522 (3.7574)  weight_decay: 0.0500 (0.0500)  time: 0.5227  data: 0.1743  max mem: 21002
Epoch: [216]  [110/312]  eta: 0:01:57  lr: 0.000819  min_lr: 0.000819  loss: 3.6196 (3.7535)  weight_decay: 0.0500 (0.0500)  time: 0.5060  data: 0.1568  max mem: 21002
Epoch: [216]  [120/312]  eta: 0:01:53  lr: 0.000818  min_lr: 0.000818  loss: 3.7999 (3.7633)  weight_decay: 0.0500 (0.0500)  time: 0.4913  data: 0.1309  max mem: 21002
Epoch: [216]  [130/312]  eta: 0:01:48  lr: 0.000818  min_lr: 0.000818  loss: 3.7889 (3.7596)  weight_decay: 0.0500 (0.0500)  time: 0.6765  data: 0.2310  max mem: 21002
Epoch: [216]  [140/312]  eta: 0:01:39  lr: 0.000817  min_lr: 0.000817  loss: 3.7598 (3.7493)  weight_decay: 0.0500 (0.0500)  time: 0.4797  data: 0.1070  max mem: 21002
Epoch: [216]  [150/312]  eta: 0:01:34  lr: 0.000817  min_lr: 0.000817  loss: 3.8187 (3.7515)  weight_decay: 0.0500 (0.0500)  time: 0.5056  data: 0.1273  max mem: 21002
Epoch: [216]  [160/312]  eta: 0:01:28  lr: 0.000816  min_lr: 0.000816  loss: 3.8677 (3.7493)  weight_decay: 0.0500 (0.0500)  time: 0.6408  data: 0.1485  max mem: 21002
Epoch: [216]  [170/312]  eta: 0:01:22  lr: 0.000815  min_lr: 0.000815  loss: 3.8143 (3.7481)  weight_decay: 0.0500 (0.0500)  time: 0.5261  data: 0.1261  max mem: 21002
Epoch: [216]  [180/312]  eta: 0:01:16  lr: 0.000815  min_lr: 0.000815  loss: 3.6266 (3.7339)  weight_decay: 0.0500 (0.0500)  time: 0.5562  data: 0.1745  max mem: 21002
Epoch: [216]  [190/312]  eta: 0:01:08  lr: 0.000814  min_lr: 0.000814  loss: 3.6870 (3.7302)  weight_decay: 0.0500 (0.0500)  time: 0.4592  data: 0.0764  max mem: 21002
Epoch: [216]  [200/312]  eta: 0:01:04  lr: 0.000814  min_lr: 0.000814  loss: 3.6760 (3.7262)  weight_decay: 0.0500 (0.0500)  time: 0.4931  data: 0.1114  max mem: 21002
Epoch: [216]  [210/312]  eta: 0:00:58  lr: 0.000813  min_lr: 0.000813  loss: 4.0298 (3.7351)  weight_decay: 0.0500 (0.0500)  time: 0.6783  data: 0.1674  max mem: 21002
Epoch: [216]  [220/312]  eta: 0:00:52  lr: 0.000812  min_lr: 0.000812  loss: 4.0298 (3.7402)  weight_decay: 0.0500 (0.0500)  time: 0.4992  data: 0.0846  max mem: 21002
Epoch: [216]  [230/312]  eta: 0:00:46  lr: 0.000812  min_lr: 0.000812  loss: 3.7659 (3.7273)  weight_decay: 0.0500 (0.0500)  time: 0.5143  data: 0.1016  max mem: 21002
Epoch: [216]  [240/312]  eta: 0:00:40  lr: 0.000811  min_lr: 0.000811  loss: 3.5625 (3.7324)  weight_decay: 0.0500 (0.0500)  time: 0.5963  data: 0.0738  max mem: 21002
Epoch: [216]  [250/312]  eta: 0:00:34  lr: 0.000811  min_lr: 0.000811  loss: 4.0168 (3.7350)  weight_decay: 0.0500 (0.0500)  time: 0.4397  data: 0.0445  max mem: 21002
Epoch: [216]  [260/312]  eta: 0:00:29  lr: 0.000810  min_lr: 0.000810  loss: 4.0352 (3.7393)  weight_decay: 0.0500 (0.0500)  time: 0.5478  data: 0.1247  max mem: 21002
Epoch: [216]  [270/312]  eta: 0:00:23  lr: 0.000810  min_lr: 0.000810  loss: 3.8781 (3.7368)  weight_decay: 0.0500 (0.0500)  time: 0.5045  data: 0.0811  max mem: 21002
Epoch: [216]  [280/312]  eta: 0:00:17  lr: 0.000809  min_lr: 0.000809  loss: 3.7501 (3.7367)  weight_decay: 0.0500 (0.0500)  time: 0.4785  data: 0.0747  max mem: 21002
Epoch: [216]  [290/312]  eta: 0:00:12  lr: 0.000808  min_lr: 0.000808  loss: 3.6002 (3.7270)  weight_decay: 0.0500 (0.0500)  time: 0.6369  data: 0.1065  max mem: 21002
Epoch: [216]  [300/312]  eta: 0:00:06  lr: 0.000808  min_lr: 0.000808  loss: 3.5848 (3.7261)  weight_decay: 0.0500 (0.0500)  time: 0.4415  data: 0.0323  max mem: 21002
Epoch: [216]  [310/312]  eta: 0:00:01  lr: 0.000807  min_lr: 0.000807  loss: 3.8955 (3.7328)  weight_decay: 0.0500 (0.0500)  time: 0.2921  data: 0.0106  max mem: 21002
Epoch: [216]  [311/312]  eta: 0:00:00  lr: 0.000807  min_lr: 0.000807  loss: 3.8955 (3.7318)  weight_decay: 0.0500 (0.0500)  time: 0.2919  data: 0.0106  max mem: 21002
Epoch: [216] Total time: 0:02:50 (0.5472 s / it)
Averaged stats: lr: 0.000807  min_lr: 0.000807  loss: 3.8955 (3.7334)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.3779 (1.3779)  acc1: 75.2604 (75.2604)  acc5: 92.5781 (92.5781)  time: 8.5747  data: 8.4558  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.8140 (1.6745)  acc1: 64.4531 (67.0560)  acc5: 86.9792 (87.8720)  time: 1.0360  data: 0.9396  max mem: 21002
Test: Total time: 0:00:09 (1.0529 s / it)
* Acc@1 67.160 Acc@5 88.188 loss 1.674
Accuracy of the model on the 50000 test images: 67.2%
Max accuracy: 67.16%
Epoch: [217]  [  0/312]  eta: 0:57:33  lr: 0.000807  min_lr: 0.000807  loss: 3.2826 (3.2826)  weight_decay: 0.0500 (0.0500)  time: 11.0698  data: 8.6629  max mem: 21002
Epoch: [217]  [ 10/312]  eta: 0:07:20  lr: 0.000807  min_lr: 0.000807  loss: 3.6709 (3.5904)  weight_decay: 0.0500 (0.0500)  time: 1.4574  data: 0.9778  max mem: 21002
Epoch: [217]  [ 20/312]  eta: 0:05:28  lr: 0.000806  min_lr: 0.000806  loss: 3.7431 (3.6583)  weight_decay: 0.0500 (0.0500)  time: 0.6263  data: 0.2433  max mem: 21002
Epoch: [217]  [ 30/312]  eta: 0:04:01  lr: 0.000805  min_lr: 0.000805  loss: 3.7804 (3.7434)  weight_decay: 0.0500 (0.0500)  time: 0.5257  data: 0.1389  max mem: 21002
Epoch: [217]  [ 40/312]  eta: 0:03:32  lr: 0.000805  min_lr: 0.000805  loss: 3.8656 (3.7703)  weight_decay: 0.0500 (0.0500)  time: 0.4197  data: 0.0804  max mem: 21002
Epoch: [217]  [ 50/312]  eta: 0:03:15  lr: 0.000804  min_lr: 0.000804  loss: 4.0466 (3.8161)  weight_decay: 0.0500 (0.0500)  time: 0.5722  data: 0.1722  max mem: 21002
Epoch: [217]  [ 60/312]  eta: 0:02:48  lr: 0.000804  min_lr: 0.000804  loss: 4.0751 (3.8180)  weight_decay: 0.0500 (0.0500)  time: 0.4448  data: 0.0938  max mem: 21002
Epoch: [217]  [ 70/312]  eta: 0:02:46  lr: 0.000803  min_lr: 0.000803  loss: 3.8852 (3.7985)  weight_decay: 0.0500 (0.0500)  time: 0.5437  data: 0.1460  max mem: 21002
Epoch: [217]  [ 80/312]  eta: 0:02:32  lr: 0.000803  min_lr: 0.000803  loss: 3.9105 (3.8232)  weight_decay: 0.0500 (0.0500)  time: 0.6145  data: 0.1448  max mem: 21002
Epoch: [217]  [ 90/312]  eta: 0:02:21  lr: 0.000802  min_lr: 0.000802  loss: 3.9326 (3.8262)  weight_decay: 0.0500 (0.0500)  time: 0.4619  data: 0.1046  max mem: 21002
Epoch: [217]  [100/312]  eta: 0:02:15  lr: 0.000801  min_lr: 0.000801  loss: 3.8182 (3.8287)  weight_decay: 0.0500 (0.0500)  time: 0.5787  data: 0.2030  max mem: 21002
Epoch: [217]  [110/312]  eta: 0:02:03  lr: 0.000801  min_lr: 0.000801  loss: 3.8847 (3.8244)  weight_decay: 0.0500 (0.0500)  time: 0.4782  data: 0.0993  max mem: 21002
Epoch: [217]  [120/312]  eta: 0:01:58  lr: 0.000800  min_lr: 0.000800  loss: 3.8039 (3.8022)  weight_decay: 0.0500 (0.0500)  time: 0.4823  data: 0.1255  max mem: 21002
Epoch: [217]  [130/312]  eta: 0:01:53  lr: 0.000800  min_lr: 0.000800  loss: 3.7975 (3.8108)  weight_decay: 0.0500 (0.0500)  time: 0.6956  data: 0.2303  max mem: 21002
Epoch: [217]  [140/312]  eta: 0:01:43  lr: 0.000799  min_lr: 0.000799  loss: 3.8938 (3.8132)  weight_decay: 0.0500 (0.0500)  time: 0.5036  data: 0.1054  max mem: 21002
Epoch: [217]  [150/312]  eta: 0:01:37  lr: 0.000799  min_lr: 0.000799  loss: 3.8144 (3.8015)  weight_decay: 0.0500 (0.0500)  time: 0.4729  data: 0.0827  max mem: 21002
Epoch: [217]  [160/312]  eta: 0:01:30  lr: 0.000798  min_lr: 0.000798  loss: 3.6448 (3.7954)  weight_decay: 0.0500 (0.0500)  time: 0.5835  data: 0.0892  max mem: 21002
Epoch: [217]  [170/312]  eta: 0:01:23  lr: 0.000797  min_lr: 0.000797  loss: 3.7254 (3.7874)  weight_decay: 0.0500 (0.0500)  time: 0.4826  data: 0.0917  max mem: 21002
Epoch: [217]  [180/312]  eta: 0:01:18  lr: 0.000797  min_lr: 0.000797  loss: 3.5791 (3.7730)  weight_decay: 0.0500 (0.0500)  time: 0.5906  data: 0.1960  max mem: 21002
Epoch: [217]  [190/312]  eta: 0:01:10  lr: 0.000796  min_lr: 0.000796  loss: 3.7884 (3.7861)  weight_decay: 0.0500 (0.0500)  time: 0.5128  data: 0.1117  max mem: 21002
Epoch: [217]  [200/312]  eta: 0:01:05  lr: 0.000796  min_lr: 0.000796  loss: 4.0049 (3.7845)  weight_decay: 0.0500 (0.0500)  time: 0.4361  data: 0.0732  max mem: 21002
Epoch: [217]  [210/312]  eta: 0:00:59  lr: 0.000795  min_lr: 0.000795  loss: 3.9975 (3.7857)  weight_decay: 0.0500 (0.0500)  time: 0.6465  data: 0.1961  max mem: 21002
Epoch: [217]  [220/312]  eta: 0:00:52  lr: 0.000794  min_lr: 0.000794  loss: 3.9541 (3.7936)  weight_decay: 0.0500 (0.0500)  time: 0.5047  data: 0.1239  max mem: 21002
Epoch: [217]  [230/312]  eta: 0:00:47  lr: 0.000794  min_lr: 0.000794  loss: 3.8023 (3.7869)  weight_decay: 0.0500 (0.0500)  time: 0.4828  data: 0.1213  max mem: 21002
Epoch: [217]  [240/312]  eta: 0:00:41  lr: 0.000793  min_lr: 0.000793  loss: 3.7442 (3.7805)  weight_decay: 0.0500 (0.0500)  time: 0.5649  data: 0.1270  max mem: 21002
Epoch: [217]  [250/312]  eta: 0:00:35  lr: 0.000793  min_lr: 0.000793  loss: 3.6509 (3.7709)  weight_decay: 0.0500 (0.0500)  time: 0.4766  data: 0.1138  max mem: 21002
Epoch: [217]  [260/312]  eta: 0:00:29  lr: 0.000792  min_lr: 0.000792  loss: 3.7515 (3.7699)  weight_decay: 0.0500 (0.0500)  time: 0.6152  data: 0.2293  max mem: 21002
Epoch: [217]  [270/312]  eta: 0:00:23  lr: 0.000792  min_lr: 0.000792  loss: 4.0591 (3.7869)  weight_decay: 0.0500 (0.0500)  time: 0.5099  data: 0.1222  max mem: 21002
Epoch: [217]  [280/312]  eta: 0:00:18  lr: 0.000791  min_lr: 0.000791  loss: 4.0591 (3.7952)  weight_decay: 0.0500 (0.0500)  time: 0.4756  data: 0.1113  max mem: 21002
Epoch: [217]  [290/312]  eta: 0:00:12  lr: 0.000790  min_lr: 0.000790  loss: 3.8697 (3.7828)  weight_decay: 0.0500 (0.0500)  time: 0.5472  data: 0.1688  max mem: 21002
Epoch: [217]  [300/312]  eta: 0:00:06  lr: 0.000790  min_lr: 0.000790  loss: 3.8362 (3.7872)  weight_decay: 0.0500 (0.0500)  time: 0.3552  data: 0.0580  max mem: 21002
Epoch: [217]  [310/312]  eta: 0:00:01  lr: 0.000789  min_lr: 0.000789  loss: 3.8362 (3.7830)  weight_decay: 0.0500 (0.0500)  time: 0.2774  data: 0.0001  max mem: 21002
Epoch: [217]  [311/312]  eta: 0:00:00  lr: 0.000789  min_lr: 0.000789  loss: 3.8534 (3.7857)  weight_decay: 0.0500 (0.0500)  time: 0.2773  data: 0.0001  max mem: 21002
Epoch: [217] Total time: 0:02:51 (0.5486 s / it)
Averaged stats: lr: 0.000789  min_lr: 0.000789  loss: 3.8534 (3.7276)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.4821 (1.4821)  acc1: 75.3906 (75.3906)  acc5: 91.9271 (91.9271)  time: 8.3526  data: 8.2336  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.8963 (1.7883)  acc1: 66.0156 (66.8320)  acc5: 86.8490 (88.0000)  time: 1.0282  data: 0.9267  max mem: 21002
Test: Total time: 0:00:09 (1.0382 s / it)
* Acc@1 66.312 Acc@5 87.780 loss 1.799
Accuracy of the model on the 50000 test images: 66.3%
Max accuracy: 67.16%
Epoch: [218]  [  0/312]  eta: 0:57:53  lr: 0.000789  min_lr: 0.000789  loss: 3.6217 (3.6217)  weight_decay: 0.0500 (0.0500)  time: 11.1325  data: 10.3290  max mem: 21002
Epoch: [218]  [ 10/312]  eta: 0:07:46  lr: 0.000789  min_lr: 0.000789  loss: 4.1683 (4.1024)  weight_decay: 0.0500 (0.0500)  time: 1.5458  data: 1.0141  max mem: 21002
Epoch: [218]  [ 20/312]  eta: 0:05:09  lr: 0.000788  min_lr: 0.000788  loss: 3.7536 (3.7573)  weight_decay: 0.0500 (0.0500)  time: 0.5559  data: 0.0913  max mem: 21002
Epoch: [218]  [ 30/312]  eta: 0:03:49  lr: 0.000788  min_lr: 0.000788  loss: 3.5562 (3.7324)  weight_decay: 0.0500 (0.0500)  time: 0.4113  data: 0.0543  max mem: 21002
Epoch: [218]  [ 40/312]  eta: 0:03:20  lr: 0.000787  min_lr: 0.000787  loss: 3.6544 (3.7473)  weight_decay: 0.0500 (0.0500)  time: 0.3972  data: 0.0936  max mem: 21002
Epoch: [218]  [ 50/312]  eta: 0:03:04  lr: 0.000786  min_lr: 0.000786  loss: 3.8696 (3.7904)  weight_decay: 0.0500 (0.0500)  time: 0.5311  data: 0.2261  max mem: 21002
Epoch: [218]  [ 60/312]  eta: 0:02:42  lr: 0.000786  min_lr: 0.000786  loss: 3.9543 (3.7946)  weight_decay: 0.0500 (0.0500)  time: 0.4578  data: 0.1589  max mem: 21002
Epoch: [218]  [ 70/312]  eta: 0:02:37  lr: 0.000785  min_lr: 0.000785  loss: 3.8594 (3.7889)  weight_decay: 0.0500 (0.0500)  time: 0.5197  data: 0.2132  max mem: 21002
Epoch: [218]  [ 80/312]  eta: 0:02:29  lr: 0.000785  min_lr: 0.000785  loss: 3.6603 (3.7482)  weight_decay: 0.0500 (0.0500)  time: 0.6477  data: 0.3082  max mem: 21002
Epoch: [218]  [ 90/312]  eta: 0:02:17  lr: 0.000784  min_lr: 0.000784  loss: 3.2657 (3.7240)  weight_decay: 0.0500 (0.0500)  time: 0.5076  data: 0.1804  max mem: 21002
Epoch: [218]  [100/312]  eta: 0:02:09  lr: 0.000784  min_lr: 0.000784  loss: 3.6686 (3.7345)  weight_decay: 0.0500 (0.0500)  time: 0.4709  data: 0.1418  max mem: 21002
Epoch: [218]  [110/312]  eta: 0:02:00  lr: 0.000783  min_lr: 0.000783  loss: 3.5995 (3.7017)  weight_decay: 0.0500 (0.0500)  time: 0.4919  data: 0.1499  max mem: 21002
Epoch: [218]  [120/312]  eta: 0:01:56  lr: 0.000782  min_lr: 0.000782  loss: 3.5995 (3.7178)  weight_decay: 0.0500 (0.0500)  time: 0.5707  data: 0.2013  max mem: 21002
Epoch: [218]  [130/312]  eta: 0:01:49  lr: 0.000782  min_lr: 0.000782  loss: 3.8391 (3.7115)  weight_decay: 0.0500 (0.0500)  time: 0.6385  data: 0.2682  max mem: 21002
Epoch: [218]  [140/312]  eta: 0:01:39  lr: 0.000781  min_lr: 0.000781  loss: 3.6658 (3.7061)  weight_decay: 0.0500 (0.0500)  time: 0.4412  data: 0.1446  max mem: 21002
Epoch: [218]  [150/312]  eta: 0:01:36  lr: 0.000781  min_lr: 0.000781  loss: 3.6658 (3.7034)  weight_decay: 0.0500 (0.0500)  time: 0.5703  data: 0.1941  max mem: 21002
Epoch: [218]  [160/312]  eta: 0:01:30  lr: 0.000780  min_lr: 0.000780  loss: 3.6343 (3.7056)  weight_decay: 0.0500 (0.0500)  time: 0.6784  data: 0.2994  max mem: 21002
Epoch: [218]  [170/312]  eta: 0:01:21  lr: 0.000780  min_lr: 0.000780  loss: 3.5797 (3.6976)  weight_decay: 0.0500 (0.0500)  time: 0.4039  data: 0.1119  max mem: 21002
Epoch: [218]  [180/312]  eta: 0:01:16  lr: 0.000779  min_lr: 0.000779  loss: 3.9611 (3.7113)  weight_decay: 0.0500 (0.0500)  time: 0.5003  data: 0.1649  max mem: 21002
Epoch: [218]  [190/312]  eta: 0:01:09  lr: 0.000778  min_lr: 0.000778  loss: 3.9888 (3.7122)  weight_decay: 0.0500 (0.0500)  time: 0.5108  data: 0.1824  max mem: 21002
Epoch: [218]  [200/312]  eta: 0:01:03  lr: 0.000778  min_lr: 0.000778  loss: 3.6755 (3.7039)  weight_decay: 0.0500 (0.0500)  time: 0.4636  data: 0.1801  max mem: 21002
Epoch: [218]  [210/312]  eta: 0:00:58  lr: 0.000777  min_lr: 0.000777  loss: 3.8145 (3.7149)  weight_decay: 0.0500 (0.0500)  time: 0.6340  data: 0.3491  max mem: 21002
Epoch: [218]  [220/312]  eta: 0:00:51  lr: 0.000777  min_lr: 0.000777  loss: 3.8411 (3.7177)  weight_decay: 0.0500 (0.0500)  time: 0.4722  data: 0.1870  max mem: 21002
Epoch: [218]  [230/312]  eta: 0:00:46  lr: 0.000776  min_lr: 0.000776  loss: 3.7786 (3.7141)  weight_decay: 0.0500 (0.0500)  time: 0.5189  data: 0.2339  max mem: 21002
Epoch: [218]  [240/312]  eta: 0:00:41  lr: 0.000776  min_lr: 0.000776  loss: 3.3324 (3.7156)  weight_decay: 0.0500 (0.0500)  time: 0.6988  data: 0.4113  max mem: 21002
Epoch: [218]  [250/312]  eta: 0:00:34  lr: 0.000775  min_lr: 0.000775  loss: 3.4315 (3.7102)  weight_decay: 0.0500 (0.0500)  time: 0.4702  data: 0.1829  max mem: 21002
Epoch: [218]  [260/312]  eta: 0:00:29  lr: 0.000774  min_lr: 0.000774  loss: 3.9417 (3.7154)  weight_decay: 0.0500 (0.0500)  time: 0.4859  data: 0.2013  max mem: 21002
Epoch: [218]  [270/312]  eta: 0:00:23  lr: 0.000774  min_lr: 0.000774  loss: 3.4087 (3.6965)  weight_decay: 0.0500 (0.0500)  time: 0.4814  data: 0.1964  max mem: 21002
Epoch: [218]  [280/312]  eta: 0:00:18  lr: 0.000773  min_lr: 0.000773  loss: 3.3332 (3.7000)  weight_decay: 0.0500 (0.0500)  time: 0.5078  data: 0.2205  max mem: 21002
Epoch: [218]  [290/312]  eta: 0:00:12  lr: 0.000773  min_lr: 0.000773  loss: 3.9417 (3.7031)  weight_decay: 0.0500 (0.0500)  time: 0.6225  data: 0.3273  max mem: 21002
Epoch: [218]  [300/312]  eta: 0:00:06  lr: 0.000772  min_lr: 0.000772  loss: 3.7260 (3.6937)  weight_decay: 0.0500 (0.0500)  time: 0.4063  data: 0.1177  max mem: 21002
Epoch: [218]  [310/312]  eta: 0:00:01  lr: 0.000772  min_lr: 0.000772  loss: 3.4911 (3.6916)  weight_decay: 0.0500 (0.0500)  time: 0.2875  data: 0.0106  max mem: 21002
Epoch: [218]  [311/312]  eta: 0:00:00  lr: 0.000772  min_lr: 0.000772  loss: 3.4911 (3.6930)  weight_decay: 0.0500 (0.0500)  time: 0.2774  data: 0.0001  max mem: 21002
Epoch: [218] Total time: 0:02:50 (0.5463 s / it)
Averaged stats: lr: 0.000772  min_lr: 0.000772  loss: 3.4911 (3.7351)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.4196 (1.4196)  acc1: 73.1771 (73.1771)  acc5: 90.4948 (90.4948)  time: 8.6461  data: 8.5271  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.6504 (1.5911)  acc1: 66.1458 (66.3520)  acc5: 87.7604 (87.9200)  time: 1.0552  data: 0.9475  max mem: 21002
Test: Total time: 0:00:09 (1.0891 s / it)
* Acc@1 66.208 Acc@5 87.704 loss 1.596
Accuracy of the model on the 50000 test images: 66.2%
Max accuracy: 67.16%
Epoch: [219]  [  0/312]  eta: 1:02:00  lr: 0.000771  min_lr: 0.000771  loss: 4.1134 (4.1134)  weight_decay: 0.0500 (0.0500)  time: 11.9253  data: 11.6397  max mem: 21002
Epoch: [219]  [ 10/312]  eta: 0:07:30  lr: 0.000771  min_lr: 0.000771  loss: 3.6651 (3.6879)  weight_decay: 0.0500 (0.0500)  time: 1.4915  data: 1.0587  max mem: 21002
Epoch: [219]  [ 20/312]  eta: 0:05:00  lr: 0.000770  min_lr: 0.000770  loss: 3.6651 (3.6695)  weight_decay: 0.0500 (0.0500)  time: 0.4832  data: 0.0765  max mem: 21002
Epoch: [219]  [ 30/312]  eta: 0:03:42  lr: 0.000770  min_lr: 0.000770  loss: 3.9580 (3.8177)  weight_decay: 0.0500 (0.0500)  time: 0.4016  data: 0.0766  max mem: 21002
Epoch: [219]  [ 40/312]  eta: 0:03:22  lr: 0.000769  min_lr: 0.000769  loss: 4.0327 (3.7583)  weight_decay: 0.0500 (0.0500)  time: 0.4437  data: 0.1564  max mem: 21002
Epoch: [219]  [ 50/312]  eta: 0:03:10  lr: 0.000769  min_lr: 0.000769  loss: 3.5465 (3.7120)  weight_decay: 0.0500 (0.0500)  time: 0.6329  data: 0.3064  max mem: 21002
Epoch: [219]  [ 60/312]  eta: 0:02:45  lr: 0.000768  min_lr: 0.000768  loss: 3.6241 (3.7207)  weight_decay: 0.0500 (0.0500)  time: 0.4746  data: 0.1508  max mem: 21002
Epoch: [219]  [ 70/312]  eta: 0:02:38  lr: 0.000767  min_lr: 0.000767  loss: 3.8712 (3.7205)  weight_decay: 0.0500 (0.0500)  time: 0.4641  data: 0.1340  max mem: 21002
Epoch: [219]  [ 80/312]  eta: 0:02:30  lr: 0.000767  min_lr: 0.000767  loss: 3.9148 (3.7603)  weight_decay: 0.0500 (0.0500)  time: 0.6328  data: 0.2742  max mem: 21002
Epoch: [219]  [ 90/312]  eta: 0:02:15  lr: 0.000766  min_lr: 0.000766  loss: 3.9103 (3.7600)  weight_decay: 0.0500 (0.0500)  time: 0.4597  data: 0.1411  max mem: 21002
Epoch: [219]  [100/312]  eta: 0:02:11  lr: 0.000766  min_lr: 0.000766  loss: 3.8744 (3.7620)  weight_decay: 0.0500 (0.0500)  time: 0.4935  data: 0.1469  max mem: 21002
Epoch: [219]  [110/312]  eta: 0:01:58  lr: 0.000765  min_lr: 0.000765  loss: 3.8761 (3.7505)  weight_decay: 0.0500 (0.0500)  time: 0.4894  data: 0.1469  max mem: 21002
Epoch: [219]  [120/312]  eta: 0:01:55  lr: 0.000765  min_lr: 0.000765  loss: 3.8691 (3.7426)  weight_decay: 0.0500 (0.0500)  time: 0.5085  data: 0.1095  max mem: 21002
Epoch: [219]  [130/312]  eta: 0:01:52  lr: 0.000764  min_lr: 0.000764  loss: 3.8870 (3.7555)  weight_decay: 0.0500 (0.0500)  time: 0.7671  data: 0.1908  max mem: 21002
Epoch: [219]  [140/312]  eta: 0:01:42  lr: 0.000764  min_lr: 0.000764  loss: 3.8870 (3.7566)  weight_decay: 0.0500 (0.0500)  time: 0.5610  data: 0.0974  max mem: 21002
Epoch: [219]  [150/312]  eta: 0:01:36  lr: 0.000763  min_lr: 0.000763  loss: 3.8484 (3.7605)  weight_decay: 0.0500 (0.0500)  time: 0.4534  data: 0.0620  max mem: 21002
Epoch: [219]  [160/312]  eta: 0:01:30  lr: 0.000762  min_lr: 0.000762  loss: 3.6927 (3.7530)  weight_decay: 0.0500 (0.0500)  time: 0.5851  data: 0.0466  max mem: 21002
Epoch: [219]  [170/312]  eta: 0:01:22  lr: 0.000762  min_lr: 0.000762  loss: 3.6671 (3.7443)  weight_decay: 0.0500 (0.0500)  time: 0.4909  data: 0.0278  max mem: 21002
Epoch: [219]  [180/312]  eta: 0:01:17  lr: 0.000761  min_lr: 0.000761  loss: 3.8125 (3.7502)  weight_decay: 0.0500 (0.0500)  time: 0.5323  data: 0.0819  max mem: 21002
Epoch: [219]  [190/312]  eta: 0:01:09  lr: 0.000761  min_lr: 0.000761  loss: 3.8066 (3.7307)  weight_decay: 0.0500 (0.0500)  time: 0.4755  data: 0.0549  max mem: 21002
Epoch: [219]  [200/312]  eta: 0:01:04  lr: 0.000760  min_lr: 0.000760  loss: 3.8055 (3.7307)  weight_decay: 0.0500 (0.0500)  time: 0.4522  data: 0.0339  max mem: 21002
Epoch: [219]  [210/312]  eta: 0:00:59  lr: 0.000760  min_lr: 0.000760  loss: 3.8633 (3.7338)  weight_decay: 0.0500 (0.0500)  time: 0.6708  data: 0.1031  max mem: 21002
Epoch: [219]  [220/312]  eta: 0:00:52  lr: 0.000759  min_lr: 0.000759  loss: 3.7135 (3.7223)  weight_decay: 0.0500 (0.0500)  time: 0.5104  data: 0.0701  max mem: 21002
Epoch: [219]  [230/312]  eta: 0:00:46  lr: 0.000758  min_lr: 0.000758  loss: 3.7852 (3.7287)  weight_decay: 0.0500 (0.0500)  time: 0.4407  data: 0.0613  max mem: 21002
Epoch: [219]  [240/312]  eta: 0:00:40  lr: 0.000758  min_lr: 0.000758  loss: 4.0046 (3.7313)  weight_decay: 0.0500 (0.0500)  time: 0.5771  data: 0.1512  max mem: 21002
Epoch: [219]  [250/312]  eta: 0:00:34  lr: 0.000757  min_lr: 0.000757  loss: 3.8944 (3.7349)  weight_decay: 0.0500 (0.0500)  time: 0.4694  data: 0.0909  max mem: 21002
Epoch: [219]  [260/312]  eta: 0:00:29  lr: 0.000757  min_lr: 0.000757  loss: 3.8829 (3.7406)  weight_decay: 0.0500 (0.0500)  time: 0.5861  data: 0.0616  max mem: 21002
Epoch: [219]  [270/312]  eta: 0:00:23  lr: 0.000756  min_lr: 0.000756  loss: 3.9442 (3.7424)  weight_decay: 0.0500 (0.0500)  time: 0.5471  data: 0.0615  max mem: 21002
Epoch: [219]  [280/312]  eta: 0:00:18  lr: 0.000756  min_lr: 0.000756  loss: 3.8810 (3.7412)  weight_decay: 0.0500 (0.0500)  time: 0.4977  data: 0.0111  max mem: 21002
Epoch: [219]  [290/312]  eta: 0:00:12  lr: 0.000755  min_lr: 0.000755  loss: 3.8542 (3.7358)  weight_decay: 0.0500 (0.0500)  time: 0.5735  data: 0.0107  max mem: 21002
Epoch: [219]  [300/312]  eta: 0:00:06  lr: 0.000755  min_lr: 0.000755  loss: 3.7603 (3.7381)  weight_decay: 0.0500 (0.0500)  time: 0.3670  data: 0.0002  max mem: 21002
Epoch: [219]  [310/312]  eta: 0:00:01  lr: 0.000754  min_lr: 0.000754  loss: 3.8155 (3.7311)  weight_decay: 0.0500 (0.0500)  time: 0.2832  data: 0.0001  max mem: 21002
Epoch: [219]  [311/312]  eta: 0:00:00  lr: 0.000754  min_lr: 0.000754  loss: 3.8155 (3.7288)  weight_decay: 0.0500 (0.0500)  time: 0.2773  data: 0.0001  max mem: 21002
Epoch: [219] Total time: 0:02:50 (0.5456 s / it)
Averaged stats: lr: 0.000754  min_lr: 0.000754  loss: 3.8155 (3.7471)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.2494 (1.2494)  acc1: 76.1719 (76.1719)  acc5: 93.3594 (93.3594)  time: 8.3720  data: 8.2544  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.5783 (1.5342)  acc1: 67.3177 (68.6880)  acc5: 89.7135 (89.2640)  time: 1.0251  data: 0.9173  max mem: 21002
Test: Total time: 0:00:09 (1.0413 s / it)
* Acc@1 68.388 Acc@5 89.290 loss 1.541
Accuracy of the model on the 50000 test images: 68.4%
Max accuracy: 68.39%
Epoch: [220]  [  0/312]  eta: 1:02:01  lr: 0.000754  min_lr: 0.000754  loss: 3.0950 (3.0950)  weight_decay: 0.0500 (0.0500)  time: 11.9286  data: 10.8271  max mem: 21002
Epoch: [220]  [ 10/312]  eta: 0:07:16  lr: 0.000753  min_lr: 0.000753  loss: 3.4853 (3.5469)  weight_decay: 0.0500 (0.0500)  time: 1.4450  data: 1.0805  max mem: 21002
Epoch: [220]  [ 20/312]  eta: 0:04:56  lr: 0.000753  min_lr: 0.000753  loss: 3.5567 (3.5812)  weight_decay: 0.0500 (0.0500)  time: 0.4698  data: 0.1648  max mem: 21002
Epoch: [220]  [ 30/312]  eta: 0:03:39  lr: 0.000752  min_lr: 0.000752  loss: 3.8175 (3.7153)  weight_decay: 0.0500 (0.0500)  time: 0.4141  data: 0.1122  max mem: 21002
Epoch: [220]  [ 40/312]  eta: 0:03:24  lr: 0.000752  min_lr: 0.000752  loss: 3.8519 (3.7030)  weight_decay: 0.0500 (0.0500)  time: 0.4725  data: 0.1864  max mem: 21002
Epoch: [220]  [ 50/312]  eta: 0:03:12  lr: 0.000751  min_lr: 0.000751  loss: 3.8519 (3.7315)  weight_decay: 0.0500 (0.0500)  time: 0.6631  data: 0.3722  max mem: 21002
Epoch: [220]  [ 60/312]  eta: 0:02:46  lr: 0.000750  min_lr: 0.000750  loss: 3.9428 (3.7597)  weight_decay: 0.0500 (0.0500)  time: 0.4748  data: 0.1864  max mem: 21002
Epoch: [220]  [ 70/312]  eta: 0:02:40  lr: 0.000750  min_lr: 0.000750  loss: 3.8298 (3.7223)  weight_decay: 0.0500 (0.0500)  time: 0.4898  data: 0.1976  max mem: 21002
Epoch: [220]  [ 80/312]  eta: 0:02:32  lr: 0.000749  min_lr: 0.000749  loss: 3.8114 (3.7366)  weight_decay: 0.0500 (0.0500)  time: 0.6538  data: 0.3356  max mem: 21002
Epoch: [220]  [ 90/312]  eta: 0:02:18  lr: 0.000749  min_lr: 0.000749  loss: 3.9461 (3.7620)  weight_decay: 0.0500 (0.0500)  time: 0.4729  data: 0.1640  max mem: 21002
Epoch: [220]  [100/312]  eta: 0:02:13  lr: 0.000748  min_lr: 0.000748  loss: 3.9623 (3.7797)  weight_decay: 0.0500 (0.0500)  time: 0.5042  data: 0.1814  max mem: 21002
Epoch: [220]  [110/312]  eta: 0:02:00  lr: 0.000748  min_lr: 0.000748  loss: 3.8097 (3.7791)  weight_decay: 0.0500 (0.0500)  time: 0.4821  data: 0.1561  max mem: 21002
Epoch: [220]  [120/312]  eta: 0:01:54  lr: 0.000747  min_lr: 0.000747  loss: 3.7262 (3.7655)  weight_decay: 0.0500 (0.0500)  time: 0.4455  data: 0.1457  max mem: 21002
Epoch: [220]  [130/312]  eta: 0:01:49  lr: 0.000747  min_lr: 0.000747  loss: 3.8027 (3.7739)  weight_decay: 0.0500 (0.0500)  time: 0.6365  data: 0.3305  max mem: 21002
Epoch: [220]  [140/312]  eta: 0:01:39  lr: 0.000746  min_lr: 0.000746  loss: 3.8065 (3.7648)  weight_decay: 0.0500 (0.0500)  time: 0.4797  data: 0.1854  max mem: 21002
Epoch: [220]  [150/312]  eta: 0:01:35  lr: 0.000745  min_lr: 0.000745  loss: 4.0072 (3.7763)  weight_decay: 0.0500 (0.0500)  time: 0.4909  data: 0.1787  max mem: 21002
Epoch: [220]  [160/312]  eta: 0:01:29  lr: 0.000745  min_lr: 0.000745  loss: 3.9642 (3.7660)  weight_decay: 0.0500 (0.0500)  time: 0.6592  data: 0.3308  max mem: 21002
Epoch: [220]  [170/312]  eta: 0:01:21  lr: 0.000744  min_lr: 0.000744  loss: 3.5211 (3.7609)  weight_decay: 0.0500 (0.0500)  time: 0.4585  data: 0.1574  max mem: 21002
Epoch: [220]  [180/312]  eta: 0:01:16  lr: 0.000744  min_lr: 0.000744  loss: 3.9256 (3.7646)  weight_decay: 0.0500 (0.0500)  time: 0.5016  data: 0.1843  max mem: 21002
Epoch: [220]  [190/312]  eta: 0:01:08  lr: 0.000743  min_lr: 0.000743  loss: 3.7528 (3.7548)  weight_decay: 0.0500 (0.0500)  time: 0.4985  data: 0.1796  max mem: 21002
Epoch: [220]  [200/312]  eta: 0:01:03  lr: 0.000743  min_lr: 0.000743  loss: 3.7102 (3.7477)  weight_decay: 0.0500 (0.0500)  time: 0.4722  data: 0.1660  max mem: 21002
Epoch: [220]  [210/312]  eta: 0:00:58  lr: 0.000742  min_lr: 0.000742  loss: 3.7972 (3.7579)  weight_decay: 0.0500 (0.0500)  time: 0.6512  data: 0.3446  max mem: 21002
Epoch: [220]  [220/312]  eta: 0:00:51  lr: 0.000742  min_lr: 0.000742  loss: 3.8447 (3.7539)  weight_decay: 0.0500 (0.0500)  time: 0.4868  data: 0.1795  max mem: 21002
Epoch: [220]  [230/312]  eta: 0:00:46  lr: 0.000741  min_lr: 0.000741  loss: 3.8767 (3.7651)  weight_decay: 0.0500 (0.0500)  time: 0.5090  data: 0.1791  max mem: 21002
Epoch: [220]  [240/312]  eta: 0:00:41  lr: 0.000740  min_lr: 0.000740  loss: 3.7989 (3.7613)  weight_decay: 0.0500 (0.0500)  time: 0.6609  data: 0.3501  max mem: 21002
Epoch: [220]  [250/312]  eta: 0:00:34  lr: 0.000740  min_lr: 0.000740  loss: 3.6708 (3.7586)  weight_decay: 0.0500 (0.0500)  time: 0.4957  data: 0.1718  max mem: 21002
Epoch: [220]  [260/312]  eta: 0:00:29  lr: 0.000739  min_lr: 0.000739  loss: 3.6755 (3.7551)  weight_decay: 0.0500 (0.0500)  time: 0.5104  data: 0.1751  max mem: 21002
Epoch: [220]  [270/312]  eta: 0:00:23  lr: 0.000739  min_lr: 0.000739  loss: 3.6755 (3.7464)  weight_decay: 0.0500 (0.0500)  time: 0.4749  data: 0.1752  max mem: 21002
Epoch: [220]  [280/312]  eta: 0:00:17  lr: 0.000738  min_lr: 0.000738  loss: 3.8109 (3.7492)  weight_decay: 0.0500 (0.0500)  time: 0.4839  data: 0.1622  max mem: 21002
Epoch: [220]  [290/312]  eta: 0:00:12  lr: 0.000738  min_lr: 0.000738  loss: 3.9916 (3.7421)  weight_decay: 0.0500 (0.0500)  time: 0.5904  data: 0.2592  max mem: 21002
Epoch: [220]  [300/312]  eta: 0:00:06  lr: 0.000737  min_lr: 0.000737  loss: 3.7821 (3.7448)  weight_decay: 0.0500 (0.0500)  time: 0.3972  data: 0.0975  max mem: 21002
Epoch: [220]  [310/312]  eta: 0:00:01  lr: 0.000736  min_lr: 0.000736  loss: 4.1349 (3.7502)  weight_decay: 0.0500 (0.0500)  time: 0.2868  data: 0.0001  max mem: 21002
Epoch: [220]  [311/312]  eta: 0:00:00  lr: 0.000736  min_lr: 0.000736  loss: 4.1349 (3.7518)  weight_decay: 0.0500 (0.0500)  time: 0.2791  data: 0.0001  max mem: 21002
Epoch: [220] Total time: 0:02:49 (0.5436 s / it)
Averaged stats: lr: 0.000736  min_lr: 0.000736  loss: 4.1349 (3.7070)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 1.4318 (1.4318)  acc1: 75.1302 (75.1302)  acc5: 92.5781 (92.5781)  time: 8.0704  data: 7.9522  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.7769 (1.6535)  acc1: 67.3177 (68.1920)  acc5: 88.0208 (88.7040)  time: 1.0107  data: 0.9134  max mem: 21002
Test: Total time: 0:00:09 (1.0250 s / it)
* Acc@1 67.432 Acc@5 88.486 loss 1.663
Accuracy of the model on the 50000 test images: 67.4%
Max accuracy: 68.39%
Epoch: [221]  [  0/312]  eta: 1:01:13  lr: 0.000736  min_lr: 0.000736  loss: 4.0058 (4.0058)  weight_decay: 0.0500 (0.0500)  time: 11.7742  data: 9.9685  max mem: 21002
Epoch: [221]  [ 10/312]  eta: 0:07:42  lr: 0.000736  min_lr: 0.000736  loss: 3.7865 (3.5934)  weight_decay: 0.0500 (0.0500)  time: 1.5303  data: 1.0057  max mem: 21002
Epoch: [221]  [ 20/312]  eta: 0:05:13  lr: 0.000735  min_lr: 0.000735  loss: 3.5111 (3.5902)  weight_decay: 0.0500 (0.0500)  time: 0.5383  data: 0.0980  max mem: 21002
Epoch: [221]  [ 30/312]  eta: 0:03:51  lr: 0.000735  min_lr: 0.000735  loss: 3.6728 (3.5918)  weight_decay: 0.0500 (0.0500)  time: 0.4300  data: 0.0436  max mem: 21002
Epoch: [221]  [ 40/312]  eta: 0:03:30  lr: 0.000734  min_lr: 0.000734  loss: 3.6845 (3.6271)  weight_decay: 0.0500 (0.0500)  time: 0.4572  data: 0.0720  max mem: 21002
Epoch: [221]  [ 50/312]  eta: 0:03:19  lr: 0.000734  min_lr: 0.000734  loss: 3.6845 (3.5901)  weight_decay: 0.0500 (0.0500)  time: 0.6695  data: 0.1727  max mem: 21002
Epoch: [221]  [ 60/312]  eta: 0:02:52  lr: 0.000733  min_lr: 0.000733  loss: 3.7453 (3.6435)  weight_decay: 0.0500 (0.0500)  time: 0.5018  data: 0.1013  max mem: 21002
Epoch: [221]  [ 70/312]  eta: 0:02:48  lr: 0.000732  min_lr: 0.000732  loss: 3.9496 (3.6724)  weight_decay: 0.0500 (0.0500)  time: 0.5296  data: 0.0390  max mem: 21002
Epoch: [221]  [ 80/312]  eta: 0:02:34  lr: 0.000732  min_lr: 0.000732  loss: 3.8604 (3.6799)  weight_decay: 0.0500 (0.0500)  time: 0.6149  data: 0.0390  max mem: 21002
Epoch: [221]  [ 90/312]  eta: 0:02:20  lr: 0.000731  min_lr: 0.000731  loss: 3.8604 (3.6862)  weight_decay: 0.0500 (0.0500)  time: 0.4088  data: 0.0163  max mem: 21002
Epoch: [221]  [100/312]  eta: 0:02:17  lr: 0.000731  min_lr: 0.000731  loss: 3.8143 (3.6835)  weight_decay: 0.0500 (0.0500)  time: 0.5719  data: 0.0502  max mem: 21002
Epoch: [221]  [110/312]  eta: 0:02:04  lr: 0.000730  min_lr: 0.000730  loss: 3.8779 (3.7108)  weight_decay: 0.0500 (0.0500)  time: 0.5407  data: 0.0346  max mem: 21002
Epoch: [221]  [120/312]  eta: 0:01:59  lr: 0.000730  min_lr: 0.000730  loss: 3.9407 (3.7254)  weight_decay: 0.0500 (0.0500)  time: 0.4824  data: 0.0008  max mem: 21002
Epoch: [221]  [130/312]  eta: 0:01:54  lr: 0.000729  min_lr: 0.000729  loss: 3.8823 (3.7212)  weight_decay: 0.0500 (0.0500)  time: 0.6829  data: 0.0268  max mem: 21002
Epoch: [221]  [140/312]  eta: 0:01:43  lr: 0.000729  min_lr: 0.000729  loss: 3.8823 (3.7298)  weight_decay: 0.0500 (0.0500)  time: 0.4947  data: 0.0275  max mem: 21002
Epoch: [221]  [150/312]  eta: 0:01:38  lr: 0.000728  min_lr: 0.000728  loss: 3.8476 (3.7286)  weight_decay: 0.0500 (0.0500)  time: 0.4954  data: 0.0289  max mem: 21002
Epoch: [221]  [160/312]  eta: 0:01:32  lr: 0.000727  min_lr: 0.000727  loss: 3.7387 (3.7221)  weight_decay: 0.0500 (0.0500)  time: 0.6618  data: 0.0298  max mem: 21002
Epoch: [221]  [170/312]  eta: 0:01:24  lr: 0.000727  min_lr: 0.000727  loss: 3.7218 (3.7189)  weight_decay: 0.0500 (0.0500)  time: 0.4894  data: 0.0331  max mem: 21002
Epoch: [221]  [180/312]  eta: 0:01:19  lr: 0.000726  min_lr: 0.000726  loss: 3.7218 (3.7143)  weight_decay: 0.0500 (0.0500)  time: 0.5560  data: 0.0734  max mem: 21002
Epoch: [221]  [190/312]  eta: 0:01:11  lr: 0.000726  min_lr: 0.000726  loss: 4.0159 (3.7312)  weight_decay: 0.0500 (0.0500)  time: 0.5223  data: 0.0427  max mem: 21002
Epoch: [221]  [200/312]  eta: 0:01:05  lr: 0.000725  min_lr: 0.000725  loss: 4.0170 (3.7323)  weight_decay: 0.0500 (0.0500)  time: 0.4180  data: 0.0007  max mem: 21002
Epoch: [221]  [210/312]  eta: 0:01:00  lr: 0.000725  min_lr: 0.000725  loss: 3.5489 (3.7300)  weight_decay: 0.0500 (0.0500)  time: 0.6191  data: 0.0017  max mem: 21002
Epoch: [221]  [220/312]  eta: 0:00:53  lr: 0.000724  min_lr: 0.000724  loss: 3.8173 (3.7339)  weight_decay: 0.0500 (0.0500)  time: 0.4913  data: 0.0017  max mem: 21002
Epoch: [221]  [230/312]  eta: 0:00:47  lr: 0.000724  min_lr: 0.000724  loss: 3.9714 (3.7435)  weight_decay: 0.0500 (0.0500)  time: 0.4930  data: 0.0010  max mem: 21002
Epoch: [221]  [240/312]  eta: 0:00:42  lr: 0.000723  min_lr: 0.000723  loss: 3.9714 (3.7422)  weight_decay: 0.0500 (0.0500)  time: 0.6938  data: 0.0081  max mem: 21002
Epoch: [221]  [250/312]  eta: 0:00:35  lr: 0.000722  min_lr: 0.000722  loss: 3.9401 (3.7455)  weight_decay: 0.0500 (0.0500)  time: 0.5261  data: 0.0430  max mem: 21002
Epoch: [221]  [260/312]  eta: 0:00:30  lr: 0.000722  min_lr: 0.000722  loss: 3.9865 (3.7528)  weight_decay: 0.0500 (0.0500)  time: 0.5499  data: 0.1149  max mem: 21002
Epoch: [221]  [270/312]  eta: 0:00:24  lr: 0.000721  min_lr: 0.000721  loss: 4.0041 (3.7676)  weight_decay: 0.0500 (0.0500)  time: 0.5141  data: 0.0798  max mem: 21002
Epoch: [221]  [280/312]  eta: 0:00:18  lr: 0.000721  min_lr: 0.000721  loss: 4.0051 (3.7748)  weight_decay: 0.0500 (0.0500)  time: 0.3945  data: 0.0052  max mem: 21002
Epoch: [221]  [290/312]  eta: 0:00:12  lr: 0.000720  min_lr: 0.000720  loss: 3.9082 (3.7659)  weight_decay: 0.0500 (0.0500)  time: 0.5278  data: 0.0204  max mem: 21002
Epoch: [221]  [300/312]  eta: 0:00:06  lr: 0.000720  min_lr: 0.000720  loss: 3.6281 (3.7602)  weight_decay: 0.0500 (0.0500)  time: 0.4137  data: 0.0155  max mem: 21002
Epoch: [221]  [310/312]  eta: 0:00:01  lr: 0.000719  min_lr: 0.000719  loss: 3.8141 (3.7627)  weight_decay: 0.0500 (0.0500)  time: 0.2768  data: 0.0001  max mem: 21002
Epoch: [221]  [311/312]  eta: 0:00:00  lr: 0.000719  min_lr: 0.000719  loss: 3.6823 (3.7612)  weight_decay: 0.0500 (0.0500)  time: 0.2768  data: 0.0001  max mem: 21002
Epoch: [221] Total time: 0:02:52 (0.5538 s / it)
Averaged stats: lr: 0.000719  min_lr: 0.000719  loss: 3.6823 (3.7550)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.4117 (1.4117)  acc1: 74.3490 (74.3490)  acc5: 92.7083 (92.7083)  time: 8.6857  data: 8.5677  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.8062 (1.6746)  acc1: 65.3646 (66.5920)  acc5: 87.7604 (87.9200)  time: 1.0483  data: 0.9520  max mem: 21002
Test: Total time: 0:00:09 (1.0647 s / it)
* Acc@1 65.820 Acc@5 87.652 loss 1.690
Accuracy of the model on the 50000 test images: 65.8%
Max accuracy: 68.39%
Epoch: [222]  [  0/312]  eta: 0:59:07  lr: 0.000719  min_lr: 0.000719  loss: 4.5315 (4.5315)  weight_decay: 0.0500 (0.0500)  time: 11.3695  data: 11.0696  max mem: 21002
Epoch: [222]  [ 10/312]  eta: 0:08:50  lr: 0.000719  min_lr: 0.000719  loss: 3.9997 (3.9675)  weight_decay: 0.0500 (0.0500)  time: 1.7570  data: 1.0070  max mem: 21002
Epoch: [222]  [ 20/312]  eta: 0:05:09  lr: 0.000718  min_lr: 0.000718  loss: 3.8269 (3.7557)  weight_decay: 0.0500 (0.0500)  time: 0.5429  data: 0.0007  max mem: 21002
Epoch: [222]  [ 30/312]  eta: 0:03:48  lr: 0.000717  min_lr: 0.000717  loss: 3.5046 (3.7285)  weight_decay: 0.0500 (0.0500)  time: 0.2898  data: 0.0016  max mem: 21002
Epoch: [222]  [ 40/312]  eta: 0:03:05  lr: 0.000717  min_lr: 0.000717  loss: 3.5046 (3.6563)  weight_decay: 0.0500 (0.0500)  time: 0.2895  data: 0.0023  max mem: 21002
Epoch: [222]  [ 50/312]  eta: 0:02:49  lr: 0.000716  min_lr: 0.000716  loss: 3.5675 (3.6400)  weight_decay: 0.0500 (0.0500)  time: 0.3923  data: 0.0494  max mem: 21002
Epoch: [222]  [ 60/312]  eta: 0:02:37  lr: 0.000716  min_lr: 0.000716  loss: 3.7577 (3.6669)  weight_decay: 0.0500 (0.0500)  time: 0.5076  data: 0.1205  max mem: 21002
Epoch: [222]  [ 70/312]  eta: 0:02:29  lr: 0.000715  min_lr: 0.000715  loss: 3.8351 (3.6687)  weight_decay: 0.0500 (0.0500)  time: 0.5386  data: 0.1208  max mem: 21002
Epoch: [222]  [ 80/312]  eta: 0:02:19  lr: 0.000715  min_lr: 0.000715  loss: 3.5722 (3.6418)  weight_decay: 0.0500 (0.0500)  time: 0.5348  data: 0.0789  max mem: 21002
Epoch: [222]  [ 90/312]  eta: 0:02:11  lr: 0.000714  min_lr: 0.000714  loss: 3.5873 (3.6493)  weight_decay: 0.0500 (0.0500)  time: 0.5145  data: 0.0675  max mem: 21002
Epoch: [222]  [100/312]  eta: 0:02:03  lr: 0.000714  min_lr: 0.000714  loss: 3.7680 (3.6587)  weight_decay: 0.0500 (0.0500)  time: 0.5099  data: 0.0654  max mem: 21002
Epoch: [222]  [110/312]  eta: 0:01:56  lr: 0.000713  min_lr: 0.000713  loss: 3.9196 (3.6596)  weight_decay: 0.0500 (0.0500)  time: 0.5018  data: 0.0754  max mem: 21002
Epoch: [222]  [120/312]  eta: 0:01:51  lr: 0.000712  min_lr: 0.000712  loss: 4.0708 (3.6926)  weight_decay: 0.0500 (0.0500)  time: 0.5543  data: 0.0876  max mem: 21002
Epoch: [222]  [130/312]  eta: 0:01:44  lr: 0.000712  min_lr: 0.000712  loss: 3.5677 (3.6774)  weight_decay: 0.0500 (0.0500)  time: 0.5643  data: 0.0411  max mem: 21002
Epoch: [222]  [140/312]  eta: 0:01:38  lr: 0.000711  min_lr: 0.000711  loss: 3.4761 (3.6795)  weight_decay: 0.0500 (0.0500)  time: 0.5448  data: 0.0641  max mem: 21002
Epoch: [222]  [150/312]  eta: 0:01:32  lr: 0.000711  min_lr: 0.000711  loss: 3.8987 (3.6903)  weight_decay: 0.0500 (0.0500)  time: 0.5435  data: 0.0899  max mem: 21002
Epoch: [222]  [160/312]  eta: 0:01:26  lr: 0.000710  min_lr: 0.000710  loss: 3.7952 (3.6933)  weight_decay: 0.0500 (0.0500)  time: 0.5306  data: 0.0513  max mem: 21002
Epoch: [222]  [170/312]  eta: 0:01:19  lr: 0.000710  min_lr: 0.000710  loss: 3.8447 (3.6963)  weight_decay: 0.0500 (0.0500)  time: 0.5058  data: 0.0586  max mem: 21002
Epoch: [222]  [180/312]  eta: 0:01:14  lr: 0.000709  min_lr: 0.000709  loss: 3.4913 (3.6764)  weight_decay: 0.0500 (0.0500)  time: 0.5277  data: 0.0530  max mem: 21002
Epoch: [222]  [190/312]  eta: 0:01:08  lr: 0.000709  min_lr: 0.000709  loss: 3.4913 (3.6748)  weight_decay: 0.0500 (0.0500)  time: 0.5516  data: 0.0446  max mem: 21002
Epoch: [222]  [200/312]  eta: 0:01:02  lr: 0.000708  min_lr: 0.000708  loss: 3.7202 (3.6796)  weight_decay: 0.0500 (0.0500)  time: 0.5246  data: 0.0459  max mem: 21002
Epoch: [222]  [210/312]  eta: 0:00:57  lr: 0.000708  min_lr: 0.000708  loss: 3.6527 (3.6683)  weight_decay: 0.0500 (0.0500)  time: 0.5656  data: 0.0343  max mem: 21002
Epoch: [222]  [220/312]  eta: 0:00:51  lr: 0.000707  min_lr: 0.000707  loss: 3.7574 (3.6756)  weight_decay: 0.0500 (0.0500)  time: 0.5437  data: 0.0439  max mem: 21002
Epoch: [222]  [230/312]  eta: 0:00:45  lr: 0.000706  min_lr: 0.000706  loss: 3.8788 (3.6748)  weight_decay: 0.0500 (0.0500)  time: 0.5158  data: 0.0469  max mem: 21002
Epoch: [222]  [240/312]  eta: 0:00:40  lr: 0.000706  min_lr: 0.000706  loss: 3.9114 (3.6823)  weight_decay: 0.0500 (0.0500)  time: 0.5530  data: 0.0621  max mem: 21002
Epoch: [222]  [250/312]  eta: 0:00:34  lr: 0.000705  min_lr: 0.000705  loss: 3.7291 (3.6763)  weight_decay: 0.0500 (0.0500)  time: 0.5385  data: 0.0635  max mem: 21002
Epoch: [222]  [260/312]  eta: 0:00:28  lr: 0.000705  min_lr: 0.000705  loss: 3.6648 (3.6807)  weight_decay: 0.0500 (0.0500)  time: 0.5078  data: 0.0330  max mem: 21002
Epoch: [222]  [270/312]  eta: 0:00:23  lr: 0.000704  min_lr: 0.000704  loss: 3.7965 (3.6904)  weight_decay: 0.0500 (0.0500)  time: 0.5362  data: 0.0472  max mem: 21002
Epoch: [222]  [280/312]  eta: 0:00:17  lr: 0.000704  min_lr: 0.000704  loss: 3.9510 (3.6911)  weight_decay: 0.0500 (0.0500)  time: 0.5584  data: 0.0910  max mem: 21002
Epoch: [222]  [290/312]  eta: 0:00:12  lr: 0.000703  min_lr: 0.000703  loss: 3.7510 (3.6877)  weight_decay: 0.0500 (0.0500)  time: 0.5499  data: 0.0635  max mem: 21002
Epoch: [222]  [300/312]  eta: 0:00:06  lr: 0.000703  min_lr: 0.000703  loss: 3.7848 (3.6896)  weight_decay: 0.0500 (0.0500)  time: 0.4856  data: 0.0421  max mem: 21002
Epoch: [222]  [310/312]  eta: 0:00:01  lr: 0.000702  min_lr: 0.000702  loss: 3.8768 (3.6984)  weight_decay: 0.0500 (0.0500)  time: 0.3578  data: 0.0372  max mem: 21002
Epoch: [222]  [311/312]  eta: 0:00:00  lr: 0.000702  min_lr: 0.000702  loss: 3.8245 (3.6955)  weight_decay: 0.0500 (0.0500)  time: 0.3574  data: 0.0372  max mem: 21002
Epoch: [222] Total time: 0:02:50 (0.5451 s / it)
Averaged stats: lr: 0.000702  min_lr: 0.000702  loss: 3.8245 (3.7327)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 1.3395 (1.3395)  acc1: 75.7812 (75.7812)  acc5: 92.9688 (92.9688)  time: 8.1930  data: 8.0742  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.6346 (1.5199)  acc1: 66.7969 (68.3040)  acc5: 88.4115 (88.5280)  time: 1.0050  data: 0.9072  max mem: 21002
Test: Total time: 0:00:09 (1.0143 s / it)
* Acc@1 67.954 Acc@5 88.550 loss 1.527
Accuracy of the model on the 50000 test images: 68.0%
Max accuracy: 68.39%
Epoch: [223]  [  0/312]  eta: 1:01:03  lr: 0.000702  min_lr: 0.000702  loss: 2.9327 (2.9327)  weight_decay: 0.0500 (0.0500)  time: 11.7422  data: 8.7269  max mem: 21002
Epoch: [223]  [ 10/312]  eta: 0:07:53  lr: 0.000701  min_lr: 0.000701  loss: 3.6464 (3.5077)  weight_decay: 0.0500 (0.0500)  time: 1.5670  data: 0.9524  max mem: 21002
Epoch: [223]  [ 20/312]  eta: 0:05:11  lr: 0.000701  min_lr: 0.000701  loss: 3.7731 (3.6127)  weight_decay: 0.0500 (0.0500)  time: 0.5323  data: 0.0877  max mem: 21002
Epoch: [223]  [ 30/312]  eta: 0:03:50  lr: 0.000700  min_lr: 0.000700  loss: 3.7731 (3.5891)  weight_decay: 0.0500 (0.0500)  time: 0.4031  data: 0.0006  max mem: 21002
Epoch: [223]  [ 40/312]  eta: 0:03:17  lr: 0.000700  min_lr: 0.000700  loss: 3.9279 (3.6418)  weight_decay: 0.0500 (0.0500)  time: 0.3717  data: 0.0176  max mem: 21002
Epoch: [223]  [ 50/312]  eta: 0:03:02  lr: 0.000699  min_lr: 0.000699  loss: 3.8880 (3.6841)  weight_decay: 0.0500 (0.0500)  time: 0.5157  data: 0.1307  max mem: 21002
Epoch: [223]  [ 60/312]  eta: 0:02:39  lr: 0.000699  min_lr: 0.000699  loss: 3.8201 (3.6840)  weight_decay: 0.0500 (0.0500)  time: 0.4343  data: 0.1137  max mem: 21002
Epoch: [223]  [ 70/312]  eta: 0:02:36  lr: 0.000698  min_lr: 0.000698  loss: 3.8638 (3.7100)  weight_decay: 0.0500 (0.0500)  time: 0.5206  data: 0.1394  max mem: 21002
Epoch: [223]  [ 80/312]  eta: 0:02:26  lr: 0.000698  min_lr: 0.000698  loss: 3.6016 (3.6804)  weight_decay: 0.0500 (0.0500)  time: 0.6351  data: 0.1979  max mem: 21002
Epoch: [223]  [ 90/312]  eta: 0:02:15  lr: 0.000697  min_lr: 0.000697  loss: 3.8170 (3.6950)  weight_decay: 0.0500 (0.0500)  time: 0.4690  data: 0.0906  max mem: 21002
Epoch: [223]  [100/312]  eta: 0:02:12  lr: 0.000696  min_lr: 0.000696  loss: 3.8721 (3.6899)  weight_decay: 0.0500 (0.0500)  time: 0.6046  data: 0.1278  max mem: 21002
Epoch: [223]  [110/312]  eta: 0:02:00  lr: 0.000696  min_lr: 0.000696  loss: 3.8721 (3.6978)  weight_decay: 0.0500 (0.0500)  time: 0.5514  data: 0.1078  max mem: 21002
Epoch: [223]  [120/312]  eta: 0:01:54  lr: 0.000695  min_lr: 0.000695  loss: 3.8067 (3.6992)  weight_decay: 0.0500 (0.0500)  time: 0.4534  data: 0.0813  max mem: 21002
Epoch: [223]  [130/312]  eta: 0:01:47  lr: 0.000695  min_lr: 0.000695  loss: 3.5614 (3.6813)  weight_decay: 0.0500 (0.0500)  time: 0.5635  data: 0.1502  max mem: 21002
Epoch: [223]  [140/312]  eta: 0:01:39  lr: 0.000694  min_lr: 0.000694  loss: 3.5614 (3.6801)  weight_decay: 0.0500 (0.0500)  time: 0.4768  data: 0.1091  max mem: 21002
Epoch: [223]  [150/312]  eta: 0:01:35  lr: 0.000694  min_lr: 0.000694  loss: 3.9844 (3.6944)  weight_decay: 0.0500 (0.0500)  time: 0.5528  data: 0.1554  max mem: 21002
Epoch: [223]  [160/312]  eta: 0:01:28  lr: 0.000693  min_lr: 0.000693  loss: 3.9563 (3.6983)  weight_decay: 0.0500 (0.0500)  time: 0.5764  data: 0.1807  max mem: 21002
Epoch: [223]  [170/312]  eta: 0:01:21  lr: 0.000693  min_lr: 0.000693  loss: 3.5830 (3.6936)  weight_decay: 0.0500 (0.0500)  time: 0.4815  data: 0.1266  max mem: 21002
Epoch: [223]  [180/312]  eta: 0:01:16  lr: 0.000692  min_lr: 0.000692  loss: 3.7949 (3.6977)  weight_decay: 0.0500 (0.0500)  time: 0.5628  data: 0.1922  max mem: 21002
Epoch: [223]  [190/312]  eta: 0:01:08  lr: 0.000692  min_lr: 0.000692  loss: 3.7949 (3.7021)  weight_decay: 0.0500 (0.0500)  time: 0.4585  data: 0.1196  max mem: 21002
Epoch: [223]  [200/312]  eta: 0:01:03  lr: 0.000691  min_lr: 0.000691  loss: 3.8879 (3.7097)  weight_decay: 0.0500 (0.0500)  time: 0.4909  data: 0.1884  max mem: 21002
Epoch: [223]  [210/312]  eta: 0:00:57  lr: 0.000690  min_lr: 0.000690  loss: 3.8879 (3.7116)  weight_decay: 0.0500 (0.0500)  time: 0.5963  data: 0.2838  max mem: 21002
Epoch: [223]  [220/312]  eta: 0:00:51  lr: 0.000690  min_lr: 0.000690  loss: 3.9596 (3.7234)  weight_decay: 0.0500 (0.0500)  time: 0.5164  data: 0.1803  max mem: 21002
Epoch: [223]  [230/312]  eta: 0:00:46  lr: 0.000689  min_lr: 0.000689  loss: 3.8643 (3.7184)  weight_decay: 0.0500 (0.0500)  time: 0.6035  data: 0.2486  max mem: 21002
Epoch: [223]  [240/312]  eta: 0:00:40  lr: 0.000689  min_lr: 0.000689  loss: 3.5398 (3.7038)  weight_decay: 0.0500 (0.0500)  time: 0.6182  data: 0.2283  max mem: 21002
Epoch: [223]  [250/312]  eta: 0:00:35  lr: 0.000688  min_lr: 0.000688  loss: 3.3931 (3.6985)  weight_decay: 0.0500 (0.0500)  time: 0.5497  data: 0.1680  max mem: 21002
Epoch: [223]  [260/312]  eta: 0:00:29  lr: 0.000688  min_lr: 0.000688  loss: 3.6589 (3.7015)  weight_decay: 0.0500 (0.0500)  time: 0.6114  data: 0.2278  max mem: 21002
Epoch: [223]  [270/312]  eta: 0:00:23  lr: 0.000687  min_lr: 0.000687  loss: 3.6640 (3.6951)  weight_decay: 0.0500 (0.0500)  time: 0.4903  data: 0.1255  max mem: 21002
Epoch: [223]  [280/312]  eta: 0:00:18  lr: 0.000687  min_lr: 0.000687  loss: 3.5531 (3.6842)  weight_decay: 0.0500 (0.0500)  time: 0.4779  data: 0.0979  max mem: 21002
Epoch: [223]  [290/312]  eta: 0:00:12  lr: 0.000686  min_lr: 0.000686  loss: 3.5531 (3.6833)  weight_decay: 0.0500 (0.0500)  time: 0.5550  data: 0.0963  max mem: 21002
Epoch: [223]  [300/312]  eta: 0:00:06  lr: 0.000686  min_lr: 0.000686  loss: 3.5585 (3.6805)  weight_decay: 0.0500 (0.0500)  time: 0.4042  data: 0.0429  max mem: 21002
Epoch: [223]  [310/312]  eta: 0:00:01  lr: 0.000685  min_lr: 0.000685  loss: 3.8189 (3.6844)  weight_decay: 0.0500 (0.0500)  time: 0.3191  data: 0.0428  max mem: 21002
Epoch: [223]  [311/312]  eta: 0:00:00  lr: 0.000685  min_lr: 0.000685  loss: 3.8189 (3.6842)  weight_decay: 0.0500 (0.0500)  time: 0.3191  data: 0.0428  max mem: 21002
Epoch: [223] Total time: 0:02:51 (0.5487 s / it)
Averaged stats: lr: 0.000685  min_lr: 0.000685  loss: 3.8189 (3.7194)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 1.2609 (1.2609)  acc1: 75.9115 (75.9115)  acc5: 94.1406 (94.1406)  time: 8.0601  data: 7.9420  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.5423 (1.4697)  acc1: 67.8385 (68.5920)  acc5: 89.4531 (89.6480)  time: 1.0306  data: 0.9328  max mem: 21002
Test: Total time: 0:00:09 (1.0439 s / it)
* Acc@1 68.976 Acc@5 89.148 loss 1.479
Accuracy of the model on the 50000 test images: 69.0%
Max accuracy: 68.98%
Epoch: [224]  [  0/312]  eta: 1:00:52  lr: 0.000685  min_lr: 0.000685  loss: 4.0429 (4.0429)  weight_decay: 0.0500 (0.0500)  time: 11.7079  data: 10.1437  max mem: 21002
Epoch: [224]  [ 10/312]  eta: 0:08:14  lr: 0.000684  min_lr: 0.000684  loss: 3.3992 (3.5829)  weight_decay: 0.0500 (0.0500)  time: 1.6390  data: 1.0362  max mem: 21002
Epoch: [224]  [ 20/312]  eta: 0:05:12  lr: 0.000684  min_lr: 0.000684  loss: 3.3992 (3.5144)  weight_decay: 0.0500 (0.0500)  time: 0.5385  data: 0.1033  max mem: 21002
Epoch: [224]  [ 30/312]  eta: 0:03:50  lr: 0.000683  min_lr: 0.000683  loss: 3.4941 (3.5116)  weight_decay: 0.0500 (0.0500)  time: 0.3670  data: 0.0409  max mem: 21002
Epoch: [224]  [ 40/312]  eta: 0:03:12  lr: 0.000683  min_lr: 0.000683  loss: 3.4079 (3.5111)  weight_decay: 0.0500 (0.0500)  time: 0.3255  data: 0.0014  max mem: 21002
Epoch: [224]  [ 50/312]  eta: 0:03:05  lr: 0.000682  min_lr: 0.000682  loss: 3.6080 (3.5442)  weight_decay: 0.0500 (0.0500)  time: 0.5390  data: 0.1073  max mem: 21002
Epoch: [224]  [ 60/312]  eta: 0:02:41  lr: 0.000682  min_lr: 0.000682  loss: 3.6591 (3.5640)  weight_decay: 0.0500 (0.0500)  time: 0.5102  data: 0.1165  max mem: 21002
Epoch: [224]  [ 70/312]  eta: 0:02:32  lr: 0.000681  min_lr: 0.000681  loss: 3.7793 (3.5874)  weight_decay: 0.0500 (0.0500)  time: 0.4278  data: 0.1274  max mem: 21002
Epoch: [224]  [ 80/312]  eta: 0:02:27  lr: 0.000681  min_lr: 0.000681  loss: 3.9200 (3.6058)  weight_decay: 0.0500 (0.0500)  time: 0.6237  data: 0.2897  max mem: 21002
Epoch: [224]  [ 90/312]  eta: 0:02:13  lr: 0.000680  min_lr: 0.000680  loss: 3.8083 (3.6321)  weight_decay: 0.0500 (0.0500)  time: 0.4949  data: 0.1741  max mem: 21002
Epoch: [224]  [100/312]  eta: 0:02:07  lr: 0.000680  min_lr: 0.000680  loss: 3.9697 (3.6550)  weight_decay: 0.0500 (0.0500)  time: 0.4546  data: 0.1391  max mem: 21002
Epoch: [224]  [110/312]  eta: 0:01:55  lr: 0.000679  min_lr: 0.000679  loss: 3.7250 (3.6298)  weight_decay: 0.0500 (0.0500)  time: 0.4518  data: 0.1379  max mem: 21002
Epoch: [224]  [120/312]  eta: 0:01:52  lr: 0.000678  min_lr: 0.000678  loss: 3.7250 (3.6564)  weight_decay: 0.0500 (0.0500)  time: 0.4924  data: 0.1530  max mem: 21002
Epoch: [224]  [130/312]  eta: 0:01:47  lr: 0.000678  min_lr: 0.000678  loss: 3.9385 (3.6687)  weight_decay: 0.0500 (0.0500)  time: 0.6780  data: 0.2892  max mem: 21002
Epoch: [224]  [140/312]  eta: 0:01:37  lr: 0.000677  min_lr: 0.000677  loss: 3.8387 (3.6720)  weight_decay: 0.0500 (0.0500)  time: 0.4731  data: 0.1368  max mem: 21002
Epoch: [224]  [150/312]  eta: 0:01:33  lr: 0.000677  min_lr: 0.000677  loss: 3.7220 (3.6765)  weight_decay: 0.0500 (0.0500)  time: 0.4788  data: 0.1357  max mem: 21002
Epoch: [224]  [160/312]  eta: 0:01:27  lr: 0.000676  min_lr: 0.000676  loss: 3.9610 (3.6867)  weight_decay: 0.0500 (0.0500)  time: 0.6491  data: 0.2518  max mem: 21002
Epoch: [224]  [170/312]  eta: 0:01:20  lr: 0.000676  min_lr: 0.000676  loss: 3.8976 (3.6802)  weight_decay: 0.0500 (0.0500)  time: 0.5162  data: 0.1486  max mem: 21002
Epoch: [224]  [180/312]  eta: 0:01:16  lr: 0.000675  min_lr: 0.000675  loss: 3.8976 (3.6890)  weight_decay: 0.0500 (0.0500)  time: 0.5615  data: 0.1388  max mem: 21002
Epoch: [224]  [190/312]  eta: 0:01:08  lr: 0.000675  min_lr: 0.000675  loss: 3.8401 (3.6849)  weight_decay: 0.0500 (0.0500)  time: 0.5013  data: 0.1068  max mem: 21002
Epoch: [224]  [200/312]  eta: 0:01:03  lr: 0.000674  min_lr: 0.000674  loss: 3.4365 (3.6669)  weight_decay: 0.0500 (0.0500)  time: 0.4551  data: 0.0784  max mem: 21002
Epoch: [224]  [210/312]  eta: 0:00:58  lr: 0.000674  min_lr: 0.000674  loss: 3.4181 (3.6623)  weight_decay: 0.0500 (0.0500)  time: 0.6550  data: 0.1721  max mem: 21002
Epoch: [224]  [220/312]  eta: 0:00:51  lr: 0.000673  min_lr: 0.000673  loss: 3.8659 (3.6660)  weight_decay: 0.0500 (0.0500)  time: 0.4863  data: 0.0943  max mem: 21002
Epoch: [224]  [230/312]  eta: 0:00:46  lr: 0.000673  min_lr: 0.000673  loss: 3.7217 (3.6580)  weight_decay: 0.0500 (0.0500)  time: 0.4927  data: 0.1045  max mem: 21002
Epoch: [224]  [240/312]  eta: 0:00:40  lr: 0.000672  min_lr: 0.000672  loss: 3.6973 (3.6634)  weight_decay: 0.0500 (0.0500)  time: 0.6448  data: 0.2066  max mem: 21002
Epoch: [224]  [250/312]  eta: 0:00:34  lr: 0.000671  min_lr: 0.000671  loss: 3.6064 (3.6546)  weight_decay: 0.0500 (0.0500)  time: 0.4452  data: 0.1029  max mem: 21002
Epoch: [224]  [260/312]  eta: 0:00:29  lr: 0.000671  min_lr: 0.000671  loss: 3.6777 (3.6571)  weight_decay: 0.0500 (0.0500)  time: 0.4983  data: 0.1285  max mem: 21002
Epoch: [224]  [270/312]  eta: 0:00:23  lr: 0.000670  min_lr: 0.000670  loss: 3.5504 (3.6468)  weight_decay: 0.0500 (0.0500)  time: 0.4923  data: 0.1284  max mem: 21002
Epoch: [224]  [280/312]  eta: 0:00:17  lr: 0.000670  min_lr: 0.000670  loss: 3.6025 (3.6494)  weight_decay: 0.0500 (0.0500)  time: 0.4913  data: 0.1191  max mem: 21002
Epoch: [224]  [290/312]  eta: 0:00:12  lr: 0.000669  min_lr: 0.000669  loss: 3.7203 (3.6438)  weight_decay: 0.0500 (0.0500)  time: 0.6882  data: 0.2210  max mem: 21002
Epoch: [224]  [300/312]  eta: 0:00:06  lr: 0.000669  min_lr: 0.000669  loss: 3.8423 (3.6516)  weight_decay: 0.0500 (0.0500)  time: 0.4789  data: 0.1023  max mem: 21002
Epoch: [224]  [310/312]  eta: 0:00:01  lr: 0.000668  min_lr: 0.000668  loss: 3.7531 (3.6529)  weight_decay: 0.0500 (0.0500)  time: 0.2795  data: 0.0001  max mem: 21002
Epoch: [224]  [311/312]  eta: 0:00:00  lr: 0.000668  min_lr: 0.000668  loss: 3.7896 (3.6534)  weight_decay: 0.0500 (0.0500)  time: 0.2794  data: 0.0001  max mem: 21002
Epoch: [224] Total time: 0:02:49 (0.5435 s / it)
Averaged stats: lr: 0.000668  min_lr: 0.000668  loss: 3.7896 (3.7116)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.3300 (1.3300)  acc1: 76.0417 (76.0417)  acc5: 92.5781 (92.5781)  time: 8.7747  data: 8.6565  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.5581 (1.5165)  acc1: 67.7083 (68.5440)  acc5: 88.8021 (89.0400)  time: 1.0605  data: 0.9619  max mem: 21002
Test: Total time: 0:00:09 (1.0766 s / it)
* Acc@1 68.956 Acc@5 89.320 loss 1.522
Accuracy of the model on the 50000 test images: 69.0%
Max accuracy: 68.98%
Epoch: [225]  [  0/312]  eta: 1:04:03  lr: 0.000668  min_lr: 0.000668  loss: 3.3149 (3.3149)  weight_decay: 0.0500 (0.0500)  time: 12.3192  data: 8.6555  max mem: 21002
Epoch: [225]  [ 10/312]  eta: 0:08:23  lr: 0.000668  min_lr: 0.000668  loss: 3.7891 (3.5525)  weight_decay: 0.0500 (0.0500)  time: 1.6685  data: 1.0796  max mem: 21002
Epoch: [225]  [ 20/312]  eta: 0:05:05  lr: 0.000667  min_lr: 0.000667  loss: 3.8153 (3.6562)  weight_decay: 0.0500 (0.0500)  time: 0.4836  data: 0.1796  max mem: 21002
Epoch: [225]  [ 30/312]  eta: 0:03:46  lr: 0.000667  min_lr: 0.000667  loss: 3.8669 (3.7088)  weight_decay: 0.0500 (0.0500)  time: 0.3263  data: 0.0194  max mem: 21002
Epoch: [225]  [ 40/312]  eta: 0:03:11  lr: 0.000666  min_lr: 0.000666  loss: 3.9443 (3.7660)  weight_decay: 0.0500 (0.0500)  time: 0.3475  data: 0.0082  max mem: 21002
Epoch: [225]  [ 50/312]  eta: 0:03:06  lr: 0.000665  min_lr: 0.000665  loss: 3.9443 (3.7730)  weight_decay: 0.0500 (0.0500)  time: 0.5755  data: 0.1161  max mem: 21002
Epoch: [225]  [ 60/312]  eta: 0:02:42  lr: 0.000665  min_lr: 0.000665  loss: 3.8029 (3.7493)  weight_decay: 0.0500 (0.0500)  time: 0.5167  data: 0.1092  max mem: 21002
Epoch: [225]  [ 70/312]  eta: 0:02:32  lr: 0.000664  min_lr: 0.000664  loss: 3.6577 (3.7474)  weight_decay: 0.0500 (0.0500)  time: 0.4231  data: 0.0698  max mem: 21002
Epoch: [225]  [ 80/312]  eta: 0:02:24  lr: 0.000664  min_lr: 0.000664  loss: 3.7811 (3.7265)  weight_decay: 0.0500 (0.0500)  time: 0.5624  data: 0.1102  max mem: 21002
Epoch: [225]  [ 90/312]  eta: 0:02:15  lr: 0.000663  min_lr: 0.000663  loss: 3.5194 (3.7105)  weight_decay: 0.0500 (0.0500)  time: 0.5250  data: 0.1371  max mem: 21002
Epoch: [225]  [100/312]  eta: 0:02:09  lr: 0.000663  min_lr: 0.000663  loss: 3.8608 (3.7284)  weight_decay: 0.0500 (0.0500)  time: 0.5508  data: 0.1801  max mem: 21002
Epoch: [225]  [110/312]  eta: 0:01:57  lr: 0.000662  min_lr: 0.000662  loss: 3.9903 (3.7470)  weight_decay: 0.0500 (0.0500)  time: 0.4543  data: 0.0844  max mem: 21002
Epoch: [225]  [120/312]  eta: 0:01:53  lr: 0.000662  min_lr: 0.000662  loss: 3.7554 (3.7395)  weight_decay: 0.0500 (0.0500)  time: 0.4852  data: 0.1083  max mem: 21002
Epoch: [225]  [130/312]  eta: 0:01:49  lr: 0.000661  min_lr: 0.000661  loss: 3.5651 (3.7265)  weight_decay: 0.0500 (0.0500)  time: 0.7169  data: 0.2272  max mem: 21002
Epoch: [225]  [140/312]  eta: 0:01:39  lr: 0.000661  min_lr: 0.000661  loss: 3.9720 (3.7268)  weight_decay: 0.0500 (0.0500)  time: 0.5218  data: 0.1201  max mem: 21002
Epoch: [225]  [150/312]  eta: 0:01:34  lr: 0.000660  min_lr: 0.000660  loss: 3.9720 (3.7289)  weight_decay: 0.0500 (0.0500)  time: 0.4783  data: 0.1066  max mem: 21002
Epoch: [225]  [160/312]  eta: 0:01:27  lr: 0.000660  min_lr: 0.000660  loss: 3.7309 (3.7225)  weight_decay: 0.0500 (0.0500)  time: 0.5672  data: 0.1065  max mem: 21002
Epoch: [225]  [170/312]  eta: 0:01:21  lr: 0.000659  min_lr: 0.000659  loss: 3.8547 (3.7252)  weight_decay: 0.0500 (0.0500)  time: 0.4872  data: 0.1099  max mem: 21002
Epoch: [225]  [180/312]  eta: 0:01:16  lr: 0.000659  min_lr: 0.000659  loss: 3.9133 (3.7312)  weight_decay: 0.0500 (0.0500)  time: 0.5866  data: 0.1854  max mem: 21002
Epoch: [225]  [190/312]  eta: 0:01:08  lr: 0.000658  min_lr: 0.000658  loss: 3.9272 (3.7346)  weight_decay: 0.0500 (0.0500)  time: 0.4775  data: 0.0763  max mem: 21002
Epoch: [225]  [200/312]  eta: 0:01:04  lr: 0.000657  min_lr: 0.000657  loss: 3.9950 (3.7444)  weight_decay: 0.0500 (0.0500)  time: 0.5172  data: 0.0766  max mem: 21002
Epoch: [225]  [210/312]  eta: 0:00:58  lr: 0.000657  min_lr: 0.000657  loss: 3.8446 (3.7439)  weight_decay: 0.0500 (0.0500)  time: 0.6725  data: 0.1189  max mem: 21002
Epoch: [225]  [220/312]  eta: 0:00:51  lr: 0.000656  min_lr: 0.000656  loss: 3.8446 (3.7479)  weight_decay: 0.0500 (0.0500)  time: 0.4465  data: 0.0432  max mem: 21002
Epoch: [225]  [230/312]  eta: 0:00:46  lr: 0.000656  min_lr: 0.000656  loss: 3.9291 (3.7437)  weight_decay: 0.0500 (0.0500)  time: 0.4953  data: 0.0725  max mem: 21002
Epoch: [225]  [240/312]  eta: 0:00:40  lr: 0.000655  min_lr: 0.000655  loss: 3.9291 (3.7521)  weight_decay: 0.0500 (0.0500)  time: 0.5410  data: 0.0727  max mem: 21002
Epoch: [225]  [250/312]  eta: 0:00:34  lr: 0.000655  min_lr: 0.000655  loss: 3.9655 (3.7537)  weight_decay: 0.0500 (0.0500)  time: 0.4815  data: 0.0575  max mem: 21002
Epoch: [225]  [260/312]  eta: 0:00:29  lr: 0.000654  min_lr: 0.000654  loss: 3.8919 (3.7546)  weight_decay: 0.0500 (0.0500)  time: 0.6436  data: 0.0987  max mem: 21002
Epoch: [225]  [270/312]  eta: 0:00:23  lr: 0.000654  min_lr: 0.000654  loss: 3.3630 (3.7398)  weight_decay: 0.0500 (0.0500)  time: 0.4985  data: 0.0421  max mem: 21002
Epoch: [225]  [280/312]  eta: 0:00:17  lr: 0.000653  min_lr: 0.000653  loss: 3.4857 (3.7403)  weight_decay: 0.0500 (0.0500)  time: 0.4863  data: 0.0393  max mem: 21002
Epoch: [225]  [290/312]  eta: 0:00:12  lr: 0.000653  min_lr: 0.000653  loss: 3.8673 (3.7504)  weight_decay: 0.0500 (0.0500)  time: 0.6444  data: 0.0551  max mem: 21002
Epoch: [225]  [300/312]  eta: 0:00:06  lr: 0.000652  min_lr: 0.000652  loss: 3.8185 (3.7408)  weight_decay: 0.0500 (0.0500)  time: 0.4436  data: 0.0198  max mem: 21002
Epoch: [225]  [310/312]  eta: 0:00:01  lr: 0.000652  min_lr: 0.000652  loss: 3.5028 (3.7344)  weight_decay: 0.0500 (0.0500)  time: 0.2806  data: 0.0036  max mem: 21002
Epoch: [225]  [311/312]  eta: 0:00:00  lr: 0.000652  min_lr: 0.000652  loss: 3.5028 (3.7345)  weight_decay: 0.0500 (0.0500)  time: 0.2804  data: 0.0036  max mem: 21002
Epoch: [225] Total time: 0:02:50 (0.5465 s / it)
Averaged stats: lr: 0.000652  min_lr: 0.000652  loss: 3.5028 (3.7252)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.2245 (1.2245)  acc1: 77.0833 (77.0833)  acc5: 92.7083 (92.7083)  time: 8.3065  data: 8.1882  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.6032 (1.4717)  acc1: 68.0990 (69.5200)  acc5: 89.4531 (89.3760)  time: 1.0391  data: 0.9376  max mem: 21002
Test: Total time: 0:00:09 (1.0501 s / it)
* Acc@1 69.258 Acc@5 89.448 loss 1.482
Accuracy of the model on the 50000 test images: 69.3%
Max accuracy: 69.26%
Epoch: [226]  [  0/312]  eta: 1:00:36  lr: 0.000651  min_lr: 0.000651  loss: 3.0311 (3.0311)  weight_decay: 0.0500 (0.0500)  time: 11.6540  data: 11.3159  max mem: 21002
Epoch: [226]  [ 10/312]  eta: 0:07:42  lr: 0.000651  min_lr: 0.000651  loss: 3.0311 (3.4373)  weight_decay: 0.0500 (0.0500)  time: 1.5318  data: 1.0293  max mem: 21002
Epoch: [226]  [ 20/312]  eta: 0:04:57  lr: 0.000650  min_lr: 0.000650  loss: 3.8635 (3.6448)  weight_decay: 0.0500 (0.0500)  time: 0.4877  data: 0.0681  max mem: 21002
Epoch: [226]  [ 30/312]  eta: 0:03:40  lr: 0.000650  min_lr: 0.000650  loss: 3.9393 (3.6510)  weight_decay: 0.0500 (0.0500)  time: 0.3717  data: 0.0682  max mem: 21002
Epoch: [226]  [ 40/312]  eta: 0:03:18  lr: 0.000649  min_lr: 0.000649  loss: 3.9060 (3.7219)  weight_decay: 0.0500 (0.0500)  time: 0.4229  data: 0.1183  max mem: 21002
Epoch: [226]  [ 50/312]  eta: 0:03:15  lr: 0.000649  min_lr: 0.000649  loss: 3.8500 (3.6858)  weight_decay: 0.0500 (0.0500)  time: 0.6850  data: 0.2592  max mem: 21002
Epoch: [226]  [ 60/312]  eta: 0:02:49  lr: 0.000648  min_lr: 0.000648  loss: 3.6779 (3.7224)  weight_decay: 0.0500 (0.0500)  time: 0.5569  data: 0.1497  max mem: 21002
Epoch: [226]  [ 70/312]  eta: 0:02:37  lr: 0.000648  min_lr: 0.000648  loss: 3.5938 (3.6726)  weight_decay: 0.0500 (0.0500)  time: 0.4159  data: 0.1013  max mem: 21002
Epoch: [226]  [ 80/312]  eta: 0:02:30  lr: 0.000647  min_lr: 0.000647  loss: 3.7147 (3.7005)  weight_decay: 0.0500 (0.0500)  time: 0.5843  data: 0.2165  max mem: 21002
Epoch: [226]  [ 90/312]  eta: 0:02:16  lr: 0.000647  min_lr: 0.000647  loss: 3.8109 (3.7039)  weight_decay: 0.0500 (0.0500)  time: 0.4885  data: 0.1515  max mem: 21002
Epoch: [226]  [100/312]  eta: 0:02:13  lr: 0.000646  min_lr: 0.000646  loss: 3.9307 (3.7071)  weight_decay: 0.0500 (0.0500)  time: 0.5529  data: 0.1597  max mem: 21002
Epoch: [226]  [110/312]  eta: 0:02:01  lr: 0.000646  min_lr: 0.000646  loss: 3.9988 (3.7253)  weight_decay: 0.0500 (0.0500)  time: 0.5281  data: 0.1323  max mem: 21002
Epoch: [226]  [120/312]  eta: 0:01:57  lr: 0.000645  min_lr: 0.000645  loss: 4.0087 (3.7394)  weight_decay: 0.0500 (0.0500)  time: 0.5026  data: 0.0908  max mem: 21002
Epoch: [226]  [130/312]  eta: 0:01:52  lr: 0.000645  min_lr: 0.000645  loss: 3.9204 (3.7575)  weight_decay: 0.0500 (0.0500)  time: 0.7002  data: 0.1606  max mem: 21002
Epoch: [226]  [140/312]  eta: 0:01:41  lr: 0.000644  min_lr: 0.000644  loss: 3.8079 (3.7381)  weight_decay: 0.0500 (0.0500)  time: 0.4906  data: 0.0750  max mem: 21002
Epoch: [226]  [150/312]  eta: 0:01:35  lr: 0.000644  min_lr: 0.000644  loss: 3.5694 (3.7403)  weight_decay: 0.0500 (0.0500)  time: 0.4370  data: 0.0562  max mem: 21002
Epoch: [226]  [160/312]  eta: 0:01:30  lr: 0.000643  min_lr: 0.000643  loss: 3.8783 (3.7358)  weight_decay: 0.0500 (0.0500)  time: 0.5871  data: 0.0525  max mem: 21002
Epoch: [226]  [170/312]  eta: 0:01:23  lr: 0.000642  min_lr: 0.000642  loss: 3.8418 (3.7446)  weight_decay: 0.0500 (0.0500)  time: 0.5370  data: 0.0962  max mem: 21002
Epoch: [226]  [180/312]  eta: 0:01:17  lr: 0.000642  min_lr: 0.000642  loss: 3.8418 (3.7427)  weight_decay: 0.0500 (0.0500)  time: 0.5203  data: 0.1442  max mem: 21002
Epoch: [226]  [190/312]  eta: 0:01:09  lr: 0.000641  min_lr: 0.000641  loss: 3.7749 (3.7288)  weight_decay: 0.0500 (0.0500)  time: 0.4262  data: 0.0498  max mem: 21002
Epoch: [226]  [200/312]  eta: 0:01:04  lr: 0.000641  min_lr: 0.000641  loss: 3.6451 (3.7210)  weight_decay: 0.0500 (0.0500)  time: 0.5250  data: 0.1180  max mem: 21002
Epoch: [226]  [210/312]  eta: 0:00:59  lr: 0.000640  min_lr: 0.000640  loss: 3.6451 (3.7092)  weight_decay: 0.0500 (0.0500)  time: 0.6857  data: 0.1968  max mem: 21002
Epoch: [226]  [220/312]  eta: 0:00:52  lr: 0.000640  min_lr: 0.000640  loss: 3.9112 (3.7182)  weight_decay: 0.0500 (0.0500)  time: 0.4483  data: 0.0799  max mem: 21002
Epoch: [226]  [230/312]  eta: 0:00:46  lr: 0.000639  min_lr: 0.000639  loss: 4.1432 (3.7390)  weight_decay: 0.0500 (0.0500)  time: 0.4945  data: 0.0763  max mem: 21002
Epoch: [226]  [240/312]  eta: 0:00:41  lr: 0.000639  min_lr: 0.000639  loss: 3.9750 (3.7377)  weight_decay: 0.0500 (0.0500)  time: 0.6260  data: 0.1160  max mem: 21002
Epoch: [226]  [250/312]  eta: 0:00:34  lr: 0.000638  min_lr: 0.000638  loss: 3.9150 (3.7394)  weight_decay: 0.0500 (0.0500)  time: 0.4685  data: 0.0917  max mem: 21002
Epoch: [226]  [260/312]  eta: 0:00:29  lr: 0.000638  min_lr: 0.000638  loss: 3.9226 (3.7456)  weight_decay: 0.0500 (0.0500)  time: 0.5401  data: 0.1611  max mem: 21002
Epoch: [226]  [270/312]  eta: 0:00:23  lr: 0.000637  min_lr: 0.000637  loss: 3.8004 (3.7418)  weight_decay: 0.0500 (0.0500)  time: 0.4890  data: 0.1099  max mem: 21002
Epoch: [226]  [280/312]  eta: 0:00:18  lr: 0.000637  min_lr: 0.000637  loss: 3.3838 (3.7319)  weight_decay: 0.0500 (0.0500)  time: 0.4834  data: 0.0976  max mem: 21002
Epoch: [226]  [290/312]  eta: 0:00:12  lr: 0.000636  min_lr: 0.000636  loss: 3.6085 (3.7274)  weight_decay: 0.0500 (0.0500)  time: 0.6120  data: 0.1588  max mem: 21002
Epoch: [226]  [300/312]  eta: 0:00:06  lr: 0.000636  min_lr: 0.000636  loss: 3.9791 (3.7330)  weight_decay: 0.0500 (0.0500)  time: 0.4103  data: 0.0616  max mem: 21002
Epoch: [226]  [310/312]  eta: 0:00:01  lr: 0.000635  min_lr: 0.000635  loss: 4.0246 (3.7385)  weight_decay: 0.0500 (0.0500)  time: 0.2775  data: 0.0001  max mem: 21002
Epoch: [226]  [311/312]  eta: 0:00:00  lr: 0.000635  min_lr: 0.000635  loss: 4.0246 (3.7403)  weight_decay: 0.0500 (0.0500)  time: 0.2774  data: 0.0001  max mem: 21002
Epoch: [226] Total time: 0:02:50 (0.5471 s / it)
Averaged stats: lr: 0.000635  min_lr: 0.000635  loss: 4.0246 (3.7006)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.2947 (1.2947)  acc1: 77.0833 (77.0833)  acc5: 93.8802 (93.8802)  time: 8.8852  data: 8.7666  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.6844 (1.5903)  acc1: 67.3177 (68.4000)  acc5: 88.2812 (89.0240)  time: 1.0723  data: 0.9741  max mem: 21002
Test: Total time: 0:00:09 (1.0849 s / it)
* Acc@1 68.402 Acc@5 89.106 loss 1.599
Accuracy of the model on the 50000 test images: 68.4%
Max accuracy: 69.26%
Epoch: [227]  [  0/312]  eta: 1:03:14  lr: 0.000635  min_lr: 0.000635  loss: 4.6478 (4.6478)  weight_decay: 0.0500 (0.0500)  time: 12.1626  data: 11.8578  max mem: 21002
Epoch: [227]  [ 10/312]  eta: 0:07:59  lr: 0.000634  min_lr: 0.000634  loss: 3.5541 (3.5816)  weight_decay: 0.0500 (0.0500)  time: 1.5885  data: 1.0785  max mem: 21002
Epoch: [227]  [ 20/312]  eta: 0:05:06  lr: 0.000634  min_lr: 0.000634  loss: 3.5541 (3.6714)  weight_decay: 0.0500 (0.0500)  time: 0.4942  data: 0.0876  max mem: 21002
Epoch: [227]  [ 30/312]  eta: 0:03:47  lr: 0.000633  min_lr: 0.000633  loss: 3.7922 (3.6571)  weight_decay: 0.0500 (0.0500)  time: 0.3742  data: 0.0877  max mem: 21002
Epoch: [227]  [ 40/312]  eta: 0:03:14  lr: 0.000633  min_lr: 0.000633  loss: 3.8779 (3.6829)  weight_decay: 0.0500 (0.0500)  time: 0.3631  data: 0.0739  max mem: 21002
Epoch: [227]  [ 50/312]  eta: 0:03:04  lr: 0.000632  min_lr: 0.000632  loss: 4.0170 (3.7642)  weight_decay: 0.0500 (0.0500)  time: 0.5469  data: 0.2606  max mem: 21002
Epoch: [227]  [ 60/312]  eta: 0:02:40  lr: 0.000632  min_lr: 0.000632  loss: 3.9220 (3.7104)  weight_decay: 0.0500 (0.0500)  time: 0.4727  data: 0.1874  max mem: 21002
Epoch: [227]  [ 70/312]  eta: 0:02:35  lr: 0.000631  min_lr: 0.000631  loss: 3.5198 (3.7024)  weight_decay: 0.0500 (0.0500)  time: 0.4919  data: 0.1607  max mem: 21002
Epoch: [227]  [ 80/312]  eta: 0:02:28  lr: 0.000631  min_lr: 0.000631  loss: 3.9061 (3.7291)  weight_decay: 0.0500 (0.0500)  time: 0.6606  data: 0.3153  max mem: 21002
Epoch: [227]  [ 90/312]  eta: 0:02:14  lr: 0.000630  min_lr: 0.000630  loss: 3.5964 (3.6977)  weight_decay: 0.0500 (0.0500)  time: 0.4775  data: 0.1552  max mem: 21002
Epoch: [227]  [100/312]  eta: 0:02:10  lr: 0.000630  min_lr: 0.000630  loss: 3.5056 (3.6908)  weight_decay: 0.0500 (0.0500)  time: 0.4981  data: 0.1503  max mem: 21002
Epoch: [227]  [110/312]  eta: 0:01:57  lr: 0.000629  min_lr: 0.000629  loss: 3.9538 (3.7311)  weight_decay: 0.0500 (0.0500)  time: 0.4771  data: 0.1508  max mem: 21002
Epoch: [227]  [120/312]  eta: 0:01:54  lr: 0.000629  min_lr: 0.000629  loss: 3.9907 (3.7342)  weight_decay: 0.0500 (0.0500)  time: 0.5049  data: 0.1461  max mem: 21002
Epoch: [227]  [130/312]  eta: 0:01:47  lr: 0.000628  min_lr: 0.000628  loss: 3.6244 (3.7358)  weight_decay: 0.0500 (0.0500)  time: 0.6405  data: 0.2653  max mem: 21002
Epoch: [227]  [140/312]  eta: 0:01:39  lr: 0.000628  min_lr: 0.000628  loss: 3.5104 (3.7080)  weight_decay: 0.0500 (0.0500)  time: 0.4918  data: 0.1213  max mem: 21002
Epoch: [227]  [150/312]  eta: 0:01:34  lr: 0.000627  min_lr: 0.000627  loss: 3.5977 (3.7093)  weight_decay: 0.0500 (0.0500)  time: 0.5445  data: 0.1102  max mem: 21002
Epoch: [227]  [160/312]  eta: 0:01:28  lr: 0.000627  min_lr: 0.000627  loss: 3.7820 (3.7109)  weight_decay: 0.0500 (0.0500)  time: 0.5874  data: 0.2102  max mem: 21002
Epoch: [227]  [170/312]  eta: 0:01:21  lr: 0.000626  min_lr: 0.000626  loss: 3.5715 (3.6950)  weight_decay: 0.0500 (0.0500)  time: 0.4993  data: 0.1015  max mem: 21002
Epoch: [227]  [180/312]  eta: 0:01:16  lr: 0.000626  min_lr: 0.000626  loss: 3.5278 (3.6952)  weight_decay: 0.0500 (0.0500)  time: 0.5840  data: 0.0516  max mem: 21002
Epoch: [227]  [190/312]  eta: 0:01:09  lr: 0.000625  min_lr: 0.000625  loss: 3.8884 (3.7058)  weight_decay: 0.0500 (0.0500)  time: 0.4851  data: 0.0516  max mem: 21002
Epoch: [227]  [200/312]  eta: 0:01:03  lr: 0.000625  min_lr: 0.000625  loss: 3.7695 (3.6923)  weight_decay: 0.0500 (0.0500)  time: 0.4758  data: 0.0500  max mem: 21002
Epoch: [227]  [210/312]  eta: 0:00:58  lr: 0.000624  min_lr: 0.000624  loss: 3.4898 (3.6807)  weight_decay: 0.0500 (0.0500)  time: 0.6514  data: 0.0835  max mem: 21002
Epoch: [227]  [220/312]  eta: 0:00:51  lr: 0.000624  min_lr: 0.000624  loss: 3.6654 (3.6816)  weight_decay: 0.0500 (0.0500)  time: 0.4961  data: 0.0538  max mem: 21002
Epoch: [227]  [230/312]  eta: 0:00:46  lr: 0.000623  min_lr: 0.000623  loss: 3.7105 (3.6790)  weight_decay: 0.0500 (0.0500)  time: 0.5178  data: 0.0536  max mem: 21002
Epoch: [227]  [240/312]  eta: 0:00:40  lr: 0.000622  min_lr: 0.000622  loss: 3.7105 (3.6822)  weight_decay: 0.0500 (0.0500)  time: 0.5738  data: 0.0693  max mem: 21002
Epoch: [227]  [250/312]  eta: 0:00:35  lr: 0.000622  min_lr: 0.000622  loss: 3.9071 (3.6856)  weight_decay: 0.0500 (0.0500)  time: 0.5053  data: 0.0457  max mem: 21002
Epoch: [227]  [260/312]  eta: 0:00:29  lr: 0.000621  min_lr: 0.000621  loss: 3.8726 (3.6870)  weight_decay: 0.0500 (0.0500)  time: 0.6099  data: 0.0163  max mem: 21002
Epoch: [227]  [270/312]  eta: 0:00:23  lr: 0.000621  min_lr: 0.000621  loss: 3.9999 (3.6995)  weight_decay: 0.0500 (0.0500)  time: 0.5053  data: 0.0324  max mem: 21002
Epoch: [227]  [280/312]  eta: 0:00:18  lr: 0.000620  min_lr: 0.000620  loss: 4.0236 (3.7049)  weight_decay: 0.0500 (0.0500)  time: 0.4985  data: 0.0311  max mem: 21002
Epoch: [227]  [290/312]  eta: 0:00:12  lr: 0.000620  min_lr: 0.000620  loss: 3.7497 (3.6968)  weight_decay: 0.0500 (0.0500)  time: 0.5444  data: 0.0141  max mem: 21002
Epoch: [227]  [300/312]  eta: 0:00:06  lr: 0.000619  min_lr: 0.000619  loss: 3.0951 (3.6797)  weight_decay: 0.0500 (0.0500)  time: 0.3959  data: 0.0185  max mem: 21002
Epoch: [227]  [310/312]  eta: 0:00:01  lr: 0.000619  min_lr: 0.000619  loss: 3.1983 (3.6790)  weight_decay: 0.0500 (0.0500)  time: 0.3184  data: 0.0095  max mem: 21002
Epoch: [227]  [311/312]  eta: 0:00:00  lr: 0.000619  min_lr: 0.000619  loss: 3.2496 (3.6782)  weight_decay: 0.0500 (0.0500)  time: 0.3183  data: 0.0095  max mem: 21002
Epoch: [227] Total time: 0:02:50 (0.5463 s / it)
Averaged stats: lr: 0.000619  min_lr: 0.000619  loss: 3.2496 (3.7229)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.2306 (1.2306)  acc1: 76.6927 (76.6927)  acc5: 93.7500 (93.7500)  time: 8.7157  data: 8.5970  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.6280 (1.4882)  acc1: 67.4479 (69.2480)  acc5: 87.6302 (89.1200)  time: 1.0538  data: 0.9553  max mem: 21002
Test: Total time: 0:00:09 (1.0693 s / it)
* Acc@1 68.862 Acc@5 89.098 loss 1.503
Accuracy of the model on the 50000 test images: 68.9%
Max accuracy: 69.26%
Epoch: [228]  [  0/312]  eta: 0:59:04  lr: 0.000619  min_lr: 0.000619  loss: 3.9566 (3.9566)  weight_decay: 0.0500 (0.0500)  time: 11.3613  data: 8.8577  max mem: 21002
Epoch: [228]  [ 10/312]  eta: 0:07:45  lr: 0.000618  min_lr: 0.000618  loss: 3.6850 (3.5527)  weight_decay: 0.0500 (0.0500)  time: 1.5413  data: 1.0508  max mem: 21002
Epoch: [228]  [ 20/312]  eta: 0:05:22  lr: 0.000618  min_lr: 0.000618  loss: 3.9731 (3.8502)  weight_decay: 0.0500 (0.0500)  time: 0.5930  data: 0.2169  max mem: 21002
Epoch: [228]  [ 30/312]  eta: 0:03:57  lr: 0.000617  min_lr: 0.000617  loss: 4.0349 (3.8674)  weight_decay: 0.0500 (0.0500)  time: 0.4598  data: 0.0821  max mem: 21002
Epoch: [228]  [ 40/312]  eta: 0:03:26  lr: 0.000617  min_lr: 0.000617  loss: 3.8137 (3.8130)  weight_decay: 0.0500 (0.0500)  time: 0.3968  data: 0.0173  max mem: 21002
Epoch: [228]  [ 50/312]  eta: 0:03:15  lr: 0.000616  min_lr: 0.000616  loss: 3.9277 (3.8239)  weight_decay: 0.0500 (0.0500)  time: 0.5975  data: 0.0777  max mem: 21002
Epoch: [228]  [ 60/312]  eta: 0:02:49  lr: 0.000616  min_lr: 0.000616  loss: 3.9202 (3.7818)  weight_decay: 0.0500 (0.0500)  time: 0.4919  data: 0.0613  max mem: 21002
Epoch: [228]  [ 70/312]  eta: 0:02:45  lr: 0.000615  min_lr: 0.000615  loss: 3.4573 (3.7435)  weight_decay: 0.0500 (0.0500)  time: 0.5192  data: 0.0856  max mem: 21002
Epoch: [228]  [ 80/312]  eta: 0:02:33  lr: 0.000615  min_lr: 0.000615  loss: 3.8068 (3.7565)  weight_decay: 0.0500 (0.0500)  time: 0.6261  data: 0.0853  max mem: 21002
Epoch: [228]  [ 90/312]  eta: 0:02:19  lr: 0.000614  min_lr: 0.000614  loss: 3.9611 (3.7558)  weight_decay: 0.0500 (0.0500)  time: 0.4308  data: 0.0381  max mem: 21002
Epoch: [228]  [100/312]  eta: 0:02:15  lr: 0.000614  min_lr: 0.000614  loss: 3.7540 (3.7440)  weight_decay: 0.0500 (0.0500)  time: 0.5506  data: 0.1262  max mem: 21002
Epoch: [228]  [110/312]  eta: 0:02:02  lr: 0.000613  min_lr: 0.000613  loss: 3.3909 (3.6882)  weight_decay: 0.0500 (0.0500)  time: 0.5166  data: 0.0896  max mem: 21002
Epoch: [228]  [120/312]  eta: 0:01:57  lr: 0.000612  min_lr: 0.000612  loss: 3.5045 (3.6967)  weight_decay: 0.0500 (0.0500)  time: 0.4667  data: 0.0682  max mem: 21002
Epoch: [228]  [130/312]  eta: 0:01:53  lr: 0.000612  min_lr: 0.000612  loss: 3.6595 (3.6854)  weight_decay: 0.0500 (0.0500)  time: 0.7076  data: 0.1583  max mem: 21002
Epoch: [228]  [140/312]  eta: 0:01:43  lr: 0.000611  min_lr: 0.000611  loss: 3.6802 (3.6882)  weight_decay: 0.0500 (0.0500)  time: 0.5324  data: 0.0916  max mem: 21002
Epoch: [228]  [150/312]  eta: 0:01:36  lr: 0.000611  min_lr: 0.000611  loss: 3.6802 (3.6760)  weight_decay: 0.0500 (0.0500)  time: 0.4187  data: 0.0312  max mem: 21002
Epoch: [228]  [160/312]  eta: 0:01:30  lr: 0.000610  min_lr: 0.000610  loss: 3.6244 (3.6780)  weight_decay: 0.0500 (0.0500)  time: 0.5523  data: 0.0312  max mem: 21002
Epoch: [228]  [170/312]  eta: 0:01:23  lr: 0.000610  min_lr: 0.000610  loss: 3.7139 (3.6776)  weight_decay: 0.0500 (0.0500)  time: 0.5286  data: 0.1030  max mem: 21002
Epoch: [228]  [180/312]  eta: 0:01:17  lr: 0.000609  min_lr: 0.000609  loss: 3.5511 (3.6683)  weight_decay: 0.0500 (0.0500)  time: 0.5600  data: 0.1761  max mem: 21002
Epoch: [228]  [190/312]  eta: 0:01:10  lr: 0.000609  min_lr: 0.000609  loss: 3.4271 (3.6640)  weight_decay: 0.0500 (0.0500)  time: 0.4569  data: 0.0746  max mem: 21002
Epoch: [228]  [200/312]  eta: 0:01:05  lr: 0.000608  min_lr: 0.000608  loss: 3.9537 (3.6671)  weight_decay: 0.0500 (0.0500)  time: 0.5288  data: 0.1152  max mem: 21002
Epoch: [228]  [210/312]  eta: 0:00:59  lr: 0.000608  min_lr: 0.000608  loss: 3.9297 (3.6708)  weight_decay: 0.0500 (0.0500)  time: 0.6824  data: 0.1890  max mem: 21002
Epoch: [228]  [220/312]  eta: 0:00:52  lr: 0.000607  min_lr: 0.000607  loss: 3.8264 (3.6683)  weight_decay: 0.0500 (0.0500)  time: 0.4426  data: 0.0753  max mem: 21002
Epoch: [228]  [230/312]  eta: 0:00:47  lr: 0.000607  min_lr: 0.000607  loss: 3.3418 (3.6515)  weight_decay: 0.0500 (0.0500)  time: 0.4997  data: 0.0990  max mem: 21002
Epoch: [228]  [240/312]  eta: 0:00:41  lr: 0.000606  min_lr: 0.000606  loss: 3.1415 (3.6345)  weight_decay: 0.0500 (0.0500)  time: 0.5829  data: 0.0990  max mem: 21002
Epoch: [228]  [250/312]  eta: 0:00:35  lr: 0.000606  min_lr: 0.000606  loss: 3.6019 (3.6389)  weight_decay: 0.0500 (0.0500)  time: 0.4564  data: 0.0889  max mem: 21002
Epoch: [228]  [260/312]  eta: 0:00:29  lr: 0.000605  min_lr: 0.000605  loss: 3.9246 (3.6459)  weight_decay: 0.0500 (0.0500)  time: 0.6081  data: 0.2032  max mem: 21002
Epoch: [228]  [270/312]  eta: 0:00:23  lr: 0.000605  min_lr: 0.000605  loss: 3.9749 (3.6500)  weight_decay: 0.0500 (0.0500)  time: 0.5234  data: 0.1166  max mem: 21002
Epoch: [228]  [280/312]  eta: 0:00:18  lr: 0.000604  min_lr: 0.000604  loss: 3.9804 (3.6502)  weight_decay: 0.0500 (0.0500)  time: 0.4547  data: 0.0850  max mem: 21002
Epoch: [228]  [290/312]  eta: 0:00:12  lr: 0.000604  min_lr: 0.000604  loss: 3.5768 (3.6451)  weight_decay: 0.0500 (0.0500)  time: 0.5672  data: 0.1381  max mem: 21002
Epoch: [228]  [300/312]  eta: 0:00:06  lr: 0.000603  min_lr: 0.000603  loss: 3.5768 (3.6527)  weight_decay: 0.0500 (0.0500)  time: 0.3965  data: 0.0552  max mem: 21002
Epoch: [228]  [310/312]  eta: 0:00:01  lr: 0.000603  min_lr: 0.000603  loss: 3.8325 (3.6565)  weight_decay: 0.0500 (0.0500)  time: 0.2773  data: 0.0001  max mem: 21002
Epoch: [228]  [311/312]  eta: 0:00:00  lr: 0.000603  min_lr: 0.000603  loss: 3.7369 (3.6533)  weight_decay: 0.0500 (0.0500)  time: 0.2772  data: 0.0001  max mem: 21002
Epoch: [228] Total time: 0:02:51 (0.5488 s / it)
Averaged stats: lr: 0.000603  min_lr: 0.000603  loss: 3.7369 (3.6957)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.3006 (1.3006)  acc1: 77.7344 (77.7344)  acc5: 93.0990 (93.0990)  time: 8.3043  data: 8.1858  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.6252 (1.5357)  acc1: 66.6667 (68.6080)  acc5: 88.6719 (89.1360)  time: 1.0241  data: 0.9218  max mem: 21002
Test: Total time: 0:00:09 (1.0381 s / it)
* Acc@1 69.098 Acc@5 89.410 loss 1.538
Accuracy of the model on the 50000 test images: 69.1%
Max accuracy: 69.26%
Epoch: [229]  [  0/312]  eta: 1:02:02  lr: 0.000603  min_lr: 0.000603  loss: 3.8982 (3.8982)  weight_decay: 0.0500 (0.0500)  time: 11.9320  data: 11.0703  max mem: 21002
Epoch: [229]  [ 10/312]  eta: 0:08:20  lr: 0.000602  min_lr: 0.000602  loss: 3.6443 (3.6174)  weight_decay: 0.0500 (0.0500)  time: 1.6570  data: 1.0989  max mem: 21002
Epoch: [229]  [ 20/312]  eta: 0:05:04  lr: 0.000602  min_lr: 0.000602  loss: 3.7695 (3.7098)  weight_decay: 0.0500 (0.0500)  time: 0.4967  data: 0.0828  max mem: 21002
Epoch: [229]  [ 30/312]  eta: 0:03:45  lr: 0.000601  min_lr: 0.000601  loss: 3.7855 (3.7474)  weight_decay: 0.0500 (0.0500)  time: 0.3257  data: 0.0322  max mem: 21002
Epoch: [229]  [ 40/312]  eta: 0:03:11  lr: 0.000601  min_lr: 0.000601  loss: 3.7063 (3.7586)  weight_decay: 0.0500 (0.0500)  time: 0.3485  data: 0.0318  max mem: 21002
Epoch: [229]  [ 50/312]  eta: 0:03:01  lr: 0.000600  min_lr: 0.000600  loss: 4.0220 (3.8166)  weight_decay: 0.0500 (0.0500)  time: 0.5261  data: 0.1806  max mem: 21002
Epoch: [229]  [ 60/312]  eta: 0:02:37  lr: 0.000600  min_lr: 0.000600  loss: 4.0220 (3.7985)  weight_decay: 0.0500 (0.0500)  time: 0.4673  data: 0.1494  max mem: 21002
Epoch: [229]  [ 70/312]  eta: 0:02:31  lr: 0.000599  min_lr: 0.000599  loss: 3.6869 (3.7934)  weight_decay: 0.0500 (0.0500)  time: 0.4635  data: 0.1622  max mem: 21002
Epoch: [229]  [ 80/312]  eta: 0:02:27  lr: 0.000598  min_lr: 0.000598  loss: 3.5308 (3.7452)  weight_decay: 0.0500 (0.0500)  time: 0.6641  data: 0.3014  max mem: 21002
Epoch: [229]  [ 90/312]  eta: 0:02:12  lr: 0.000598  min_lr: 0.000598  loss: 3.5308 (3.7255)  weight_decay: 0.0500 (0.0500)  time: 0.4939  data: 0.1419  max mem: 21002
Epoch: [229]  [100/312]  eta: 0:02:08  lr: 0.000597  min_lr: 0.000597  loss: 3.6471 (3.7104)  weight_decay: 0.0500 (0.0500)  time: 0.4931  data: 0.1316  max mem: 21002
Epoch: [229]  [110/312]  eta: 0:01:57  lr: 0.000597  min_lr: 0.000597  loss: 3.6964 (3.7106)  weight_decay: 0.0500 (0.0500)  time: 0.5012  data: 0.1297  max mem: 21002
Epoch: [229]  [120/312]  eta: 0:01:53  lr: 0.000596  min_lr: 0.000596  loss: 3.8234 (3.7130)  weight_decay: 0.0500 (0.0500)  time: 0.5173  data: 0.1405  max mem: 21002
Epoch: [229]  [130/312]  eta: 0:01:48  lr: 0.000596  min_lr: 0.000596  loss: 3.8571 (3.7326)  weight_decay: 0.0500 (0.0500)  time: 0.6696  data: 0.2484  max mem: 21002
Epoch: [229]  [140/312]  eta: 0:01:38  lr: 0.000595  min_lr: 0.000595  loss: 3.8571 (3.7320)  weight_decay: 0.0500 (0.0500)  time: 0.4537  data: 0.1087  max mem: 21002
Epoch: [229]  [150/312]  eta: 0:01:33  lr: 0.000595  min_lr: 0.000595  loss: 3.7220 (3.7255)  weight_decay: 0.0500 (0.0500)  time: 0.4801  data: 0.1321  max mem: 21002
Epoch: [229]  [160/312]  eta: 0:01:28  lr: 0.000594  min_lr: 0.000594  loss: 3.7545 (3.7114)  weight_decay: 0.0500 (0.0500)  time: 0.6649  data: 0.2328  max mem: 21002
Epoch: [229]  [170/312]  eta: 0:01:21  lr: 0.000594  min_lr: 0.000594  loss: 3.8826 (3.7228)  weight_decay: 0.0500 (0.0500)  time: 0.5160  data: 0.1426  max mem: 21002
Epoch: [229]  [180/312]  eta: 0:01:16  lr: 0.000593  min_lr: 0.000593  loss: 3.8554 (3.7152)  weight_decay: 0.0500 (0.0500)  time: 0.5099  data: 0.1219  max mem: 21002
Epoch: [229]  [190/312]  eta: 0:01:08  lr: 0.000593  min_lr: 0.000593  loss: 3.6837 (3.7109)  weight_decay: 0.0500 (0.0500)  time: 0.4756  data: 0.0885  max mem: 21002
Epoch: [229]  [200/312]  eta: 0:01:03  lr: 0.000592  min_lr: 0.000592  loss: 3.6837 (3.7038)  weight_decay: 0.0500 (0.0500)  time: 0.5021  data: 0.1265  max mem: 21002
Epoch: [229]  [210/312]  eta: 0:00:58  lr: 0.000592  min_lr: 0.000592  loss: 3.7506 (3.6998)  weight_decay: 0.0500 (0.0500)  time: 0.6688  data: 0.1873  max mem: 21002
Epoch: [229]  [220/312]  eta: 0:00:51  lr: 0.000591  min_lr: 0.000591  loss: 3.7818 (3.6920)  weight_decay: 0.0500 (0.0500)  time: 0.4662  data: 0.0738  max mem: 21002
Epoch: [229]  [230/312]  eta: 0:00:46  lr: 0.000591  min_lr: 0.000591  loss: 3.7311 (3.6847)  weight_decay: 0.0500 (0.0500)  time: 0.4838  data: 0.0732  max mem: 21002
Epoch: [229]  [240/312]  eta: 0:00:40  lr: 0.000590  min_lr: 0.000590  loss: 3.7322 (3.6865)  weight_decay: 0.0500 (0.0500)  time: 0.6422  data: 0.1011  max mem: 21002
Epoch: [229]  [250/312]  eta: 0:00:34  lr: 0.000590  min_lr: 0.000590  loss: 3.7175 (3.6786)  weight_decay: 0.0500 (0.0500)  time: 0.5058  data: 0.0881  max mem: 21002
Epoch: [229]  [260/312]  eta: 0:00:29  lr: 0.000589  min_lr: 0.000589  loss: 3.8215 (3.6863)  weight_decay: 0.0500 (0.0500)  time: 0.5174  data: 0.1241  max mem: 21002
Epoch: [229]  [270/312]  eta: 0:00:23  lr: 0.000589  min_lr: 0.000589  loss: 3.9536 (3.7017)  weight_decay: 0.0500 (0.0500)  time: 0.4798  data: 0.0856  max mem: 21002
Epoch: [229]  [280/312]  eta: 0:00:17  lr: 0.000588  min_lr: 0.000588  loss: 4.0088 (3.7045)  weight_decay: 0.0500 (0.0500)  time: 0.4813  data: 0.0733  max mem: 21002
Epoch: [229]  [290/312]  eta: 0:00:12  lr: 0.000588  min_lr: 0.000588  loss: 3.7961 (3.7091)  weight_decay: 0.0500 (0.0500)  time: 0.6334  data: 0.1238  max mem: 21002
Epoch: [229]  [300/312]  eta: 0:00:06  lr: 0.000587  min_lr: 0.000587  loss: 3.7642 (3.7118)  weight_decay: 0.0500 (0.0500)  time: 0.4600  data: 0.0674  max mem: 21002
Epoch: [229]  [310/312]  eta: 0:00:01  lr: 0.000587  min_lr: 0.000587  loss: 3.7546 (3.7110)  weight_decay: 0.0500 (0.0500)  time: 0.2915  data: 0.0001  max mem: 21002
Epoch: [229]  [311/312]  eta: 0:00:00  lr: 0.000587  min_lr: 0.000587  loss: 3.7692 (3.7122)  weight_decay: 0.0500 (0.0500)  time: 0.2914  data: 0.0001  max mem: 21002
Epoch: [229] Total time: 0:02:49 (0.5447 s / it)
Averaged stats: lr: 0.000587  min_lr: 0.000587  loss: 3.7692 (3.6838)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.3930 (1.3930)  acc1: 76.4323 (76.4323)  acc5: 93.2292 (93.2292)  time: 8.8576  data: 8.7389  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.6518 (1.5752)  acc1: 67.4479 (69.3760)  acc5: 89.7135 (89.3440)  time: 1.0704  data: 0.9711  max mem: 21002
Test: Total time: 0:00:09 (1.1058 s / it)
* Acc@1 69.004 Acc@5 89.432 loss 1.579
Accuracy of the model on the 50000 test images: 69.0%
Max accuracy: 69.26%
Epoch: [230]  [  0/312]  eta: 1:02:28  lr: 0.000587  min_lr: 0.000587  loss: 2.8009 (2.8009)  weight_decay: 0.0500 (0.0500)  time: 12.0141  data: 10.3898  max mem: 21002
Epoch: [230]  [ 10/312]  eta: 0:07:23  lr: 0.000586  min_lr: 0.000586  loss: 3.7219 (3.7025)  weight_decay: 0.0500 (0.0500)  time: 1.4683  data: 1.0609  max mem: 21002
Epoch: [230]  [ 20/312]  eta: 0:05:08  lr: 0.000586  min_lr: 0.000586  loss: 3.6408 (3.6501)  weight_decay: 0.0500 (0.0500)  time: 0.5069  data: 0.1772  max mem: 21002
Epoch: [230]  [ 30/312]  eta: 0:03:47  lr: 0.000585  min_lr: 0.000585  loss: 3.8080 (3.7654)  weight_decay: 0.0500 (0.0500)  time: 0.4451  data: 0.1140  max mem: 21002
Epoch: [230]  [ 40/312]  eta: 0:03:27  lr: 0.000585  min_lr: 0.000585  loss: 3.8968 (3.7824)  weight_decay: 0.0500 (0.0500)  time: 0.4548  data: 0.1357  max mem: 21002
Epoch: [230]  [ 50/312]  eta: 0:03:12  lr: 0.000584  min_lr: 0.000584  loss: 3.8068 (3.8183)  weight_decay: 0.0500 (0.0500)  time: 0.6176  data: 0.2833  max mem: 21002
Epoch: [230]  [ 60/312]  eta: 0:02:47  lr: 0.000584  min_lr: 0.000584  loss: 3.7870 (3.7925)  weight_decay: 0.0500 (0.0500)  time: 0.4604  data: 0.1486  max mem: 21002
Epoch: [230]  [ 70/312]  eta: 0:02:41  lr: 0.000583  min_lr: 0.000583  loss: 3.4269 (3.7547)  weight_decay: 0.0500 (0.0500)  time: 0.5059  data: 0.1254  max mem: 21002
Epoch: [230]  [ 80/312]  eta: 0:02:34  lr: 0.000583  min_lr: 0.000583  loss: 3.4108 (3.7335)  weight_decay: 0.0500 (0.0500)  time: 0.6671  data: 0.2023  max mem: 21002
Epoch: [230]  [ 90/312]  eta: 0:02:21  lr: 0.000582  min_lr: 0.000582  loss: 3.7671 (3.7387)  weight_decay: 0.0500 (0.0500)  time: 0.5256  data: 0.1274  max mem: 21002
Epoch: [230]  [100/312]  eta: 0:02:16  lr: 0.000582  min_lr: 0.000582  loss: 3.8533 (3.7370)  weight_decay: 0.0500 (0.0500)  time: 0.5523  data: 0.1139  max mem: 21002
Epoch: [230]  [110/312]  eta: 0:02:03  lr: 0.000581  min_lr: 0.000581  loss: 3.9597 (3.7505)  weight_decay: 0.0500 (0.0500)  time: 0.4918  data: 0.0682  max mem: 21002
Epoch: [230]  [120/312]  eta: 0:01:56  lr: 0.000581  min_lr: 0.000581  loss: 3.9084 (3.7442)  weight_decay: 0.0500 (0.0500)  time: 0.4247  data: 0.0550  max mem: 21002
Epoch: [230]  [130/312]  eta: 0:01:51  lr: 0.000580  min_lr: 0.000580  loss: 3.8248 (3.7282)  weight_decay: 0.0500 (0.0500)  time: 0.6041  data: 0.1408  max mem: 21002
Epoch: [230]  [140/312]  eta: 0:01:41  lr: 0.000580  min_lr: 0.000580  loss: 3.5231 (3.7165)  weight_decay: 0.0500 (0.0500)  time: 0.5108  data: 0.1284  max mem: 21002
Epoch: [230]  [150/312]  eta: 0:01:37  lr: 0.000579  min_lr: 0.000579  loss: 3.9000 (3.7363)  weight_decay: 0.0500 (0.0500)  time: 0.5583  data: 0.1535  max mem: 21002
Epoch: [230]  [160/312]  eta: 0:01:30  lr: 0.000579  min_lr: 0.000579  loss: 3.9513 (3.7484)  weight_decay: 0.0500 (0.0500)  time: 0.5981  data: 0.1337  max mem: 21002
Epoch: [230]  [170/312]  eta: 0:01:24  lr: 0.000578  min_lr: 0.000578  loss: 3.6435 (3.7167)  weight_decay: 0.0500 (0.0500)  time: 0.5036  data: 0.0888  max mem: 21002
Epoch: [230]  [180/312]  eta: 0:01:17  lr: 0.000578  min_lr: 0.000578  loss: 3.2923 (3.7080)  weight_decay: 0.0500 (0.0500)  time: 0.5446  data: 0.1044  max mem: 21002
Epoch: [230]  [190/312]  eta: 0:01:09  lr: 0.000577  min_lr: 0.000577  loss: 3.6902 (3.6989)  weight_decay: 0.0500 (0.0500)  time: 0.4078  data: 0.0388  max mem: 21002
Epoch: [230]  [200/312]  eta: 0:01:05  lr: 0.000577  min_lr: 0.000577  loss: 3.7631 (3.7063)  weight_decay: 0.0500 (0.0500)  time: 0.5304  data: 0.0906  max mem: 21002
Epoch: [230]  [210/312]  eta: 0:00:59  lr: 0.000576  min_lr: 0.000576  loss: 3.8131 (3.7129)  weight_decay: 0.0500 (0.0500)  time: 0.7087  data: 0.1088  max mem: 21002
Epoch: [230]  [220/312]  eta: 0:00:53  lr: 0.000575  min_lr: 0.000575  loss: 3.7396 (3.7104)  weight_decay: 0.0500 (0.0500)  time: 0.5204  data: 0.0733  max mem: 21002
Epoch: [230]  [230/312]  eta: 0:00:47  lr: 0.000575  min_lr: 0.000575  loss: 3.6010 (3.7055)  weight_decay: 0.0500 (0.0500)  time: 0.4791  data: 0.1066  max mem: 21002
Epoch: [230]  [240/312]  eta: 0:00:41  lr: 0.000574  min_lr: 0.000574  loss: 3.4284 (3.7002)  weight_decay: 0.0500 (0.0500)  time: 0.5062  data: 0.0624  max mem: 21002
Epoch: [230]  [250/312]  eta: 0:00:35  lr: 0.000574  min_lr: 0.000574  loss: 3.5866 (3.7059)  weight_decay: 0.0500 (0.0500)  time: 0.5149  data: 0.1107  max mem: 21002
Epoch: [230]  [260/312]  eta: 0:00:29  lr: 0.000573  min_lr: 0.000573  loss: 3.7604 (3.7053)  weight_decay: 0.0500 (0.0500)  time: 0.5905  data: 0.1769  max mem: 21002
Epoch: [230]  [270/312]  eta: 0:00:23  lr: 0.000573  min_lr: 0.000573  loss: 3.5200 (3.6939)  weight_decay: 0.0500 (0.0500)  time: 0.4448  data: 0.0723  max mem: 21002
Epoch: [230]  [280/312]  eta: 0:00:18  lr: 0.000572  min_lr: 0.000572  loss: 3.6457 (3.6980)  weight_decay: 0.0500 (0.0500)  time: 0.5397  data: 0.0854  max mem: 21002
Epoch: [230]  [290/312]  eta: 0:00:12  lr: 0.000572  min_lr: 0.000572  loss: 3.8196 (3.6998)  weight_decay: 0.0500 (0.0500)  time: 0.6202  data: 0.1067  max mem: 21002
Epoch: [230]  [300/312]  eta: 0:00:06  lr: 0.000571  min_lr: 0.000571  loss: 3.7950 (3.6999)  weight_decay: 0.0500 (0.0500)  time: 0.3736  data: 0.0297  max mem: 21002
Epoch: [230]  [310/312]  eta: 0:00:01  lr: 0.000571  min_lr: 0.000571  loss: 3.7950 (3.7028)  weight_decay: 0.0500 (0.0500)  time: 0.2845  data: 0.0080  max mem: 21002
Epoch: [230]  [311/312]  eta: 0:00:00  lr: 0.000571  min_lr: 0.000571  loss: 3.7950 (3.7043)  weight_decay: 0.0500 (0.0500)  time: 0.2844  data: 0.0080  max mem: 21002
Epoch: [230] Total time: 0:02:51 (0.5505 s / it)
Averaged stats: lr: 0.000571  min_lr: 0.000571  loss: 3.7950 (3.7020)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.3655 (1.3655)  acc1: 75.7812 (75.7812)  acc5: 94.0104 (94.0104)  time: 8.7133  data: 8.5946  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.6867 (1.6067)  acc1: 69.4010 (69.2800)  acc5: 89.4531 (89.3760)  time: 1.0530  data: 0.9550  max mem: 21002
Test: Total time: 0:00:09 (1.0742 s / it)
* Acc@1 69.002 Acc@5 89.282 loss 1.613
Accuracy of the model on the 50000 test images: 69.0%
Max accuracy: 69.26%
Epoch: [231]  [  0/312]  eta: 1:01:22  lr: 0.000571  min_lr: 0.000571  loss: 3.6878 (3.6878)  weight_decay: 0.0500 (0.0500)  time: 11.8038  data: 9.5010  max mem: 21002
Epoch: [231]  [ 10/312]  eta: 0:08:32  lr: 0.000570  min_lr: 0.000570  loss: 3.9084 (3.6213)  weight_decay: 0.0500 (0.0500)  time: 1.6981  data: 1.0794  max mem: 21002
Epoch: [231]  [ 20/312]  eta: 0:04:59  lr: 0.000570  min_lr: 0.000570  loss: 3.9084 (3.6844)  weight_decay: 0.0500 (0.0500)  time: 0.4879  data: 0.1189  max mem: 21002
Epoch: [231]  [ 30/312]  eta: 0:03:42  lr: 0.000569  min_lr: 0.000569  loss: 3.9325 (3.7336)  weight_decay: 0.0500 (0.0500)  time: 0.2886  data: 0.0013  max mem: 21002
Epoch: [231]  [ 40/312]  eta: 0:03:06  lr: 0.000569  min_lr: 0.000569  loss: 3.9247 (3.7357)  weight_decay: 0.0500 (0.0500)  time: 0.3299  data: 0.0074  max mem: 21002
Epoch: [231]  [ 50/312]  eta: 0:03:02  lr: 0.000568  min_lr: 0.000568  loss: 3.8507 (3.7894)  weight_decay: 0.0500 (0.0500)  time: 0.5504  data: 0.1627  max mem: 21002
Epoch: [231]  [ 60/312]  eta: 0:02:38  lr: 0.000568  min_lr: 0.000568  loss: 3.8744 (3.7887)  weight_decay: 0.0500 (0.0500)  time: 0.5126  data: 0.1571  max mem: 21002
Epoch: [231]  [ 70/312]  eta: 0:02:32  lr: 0.000567  min_lr: 0.000567  loss: 3.8942 (3.7772)  weight_decay: 0.0500 (0.0500)  time: 0.4661  data: 0.1260  max mem: 21002
Epoch: [231]  [ 80/312]  eta: 0:02:25  lr: 0.000567  min_lr: 0.000567  loss: 3.8942 (3.7939)  weight_decay: 0.0500 (0.0500)  time: 0.6136  data: 0.2377  max mem: 21002
Epoch: [231]  [ 90/312]  eta: 0:02:12  lr: 0.000566  min_lr: 0.000566  loss: 3.8595 (3.7780)  weight_decay: 0.0500 (0.0500)  time: 0.4677  data: 0.1300  max mem: 21002
Epoch: [231]  [100/312]  eta: 0:02:07  lr: 0.000566  min_lr: 0.000566  loss: 3.8616 (3.7865)  weight_decay: 0.0500 (0.0500)  time: 0.4991  data: 0.1709  max mem: 21002
Epoch: [231]  [110/312]  eta: 0:01:56  lr: 0.000565  min_lr: 0.000565  loss: 3.8912 (3.7859)  weight_decay: 0.0500 (0.0500)  time: 0.4970  data: 0.1637  max mem: 21002
Epoch: [231]  [120/312]  eta: 0:01:50  lr: 0.000565  min_lr: 0.000565  loss: 4.0333 (3.8144)  weight_decay: 0.0500 (0.0500)  time: 0.4586  data: 0.1445  max mem: 21002
Epoch: [231]  [130/312]  eta: 0:01:46  lr: 0.000564  min_lr: 0.000564  loss: 3.9317 (3.8053)  weight_decay: 0.0500 (0.0500)  time: 0.6442  data: 0.2915  max mem: 21002
Epoch: [231]  [140/312]  eta: 0:01:38  lr: 0.000564  min_lr: 0.000564  loss: 3.9069 (3.8188)  weight_decay: 0.0500 (0.0500)  time: 0.5335  data: 0.1762  max mem: 21002
Epoch: [231]  [150/312]  eta: 0:01:33  lr: 0.000563  min_lr: 0.000563  loss: 3.7937 (3.8056)  weight_decay: 0.0500 (0.0500)  time: 0.5185  data: 0.1502  max mem: 21002
Epoch: [231]  [160/312]  eta: 0:01:27  lr: 0.000563  min_lr: 0.000563  loss: 3.7754 (3.7998)  weight_decay: 0.0500 (0.0500)  time: 0.6197  data: 0.2659  max mem: 21002
Epoch: [231]  [170/312]  eta: 0:01:20  lr: 0.000562  min_lr: 0.000562  loss: 3.8057 (3.8063)  weight_decay: 0.0500 (0.0500)  time: 0.4662  data: 0.1676  max mem: 21002
Epoch: [231]  [180/312]  eta: 0:01:15  lr: 0.000562  min_lr: 0.000562  loss: 3.9822 (3.8043)  weight_decay: 0.0500 (0.0500)  time: 0.5225  data: 0.1371  max mem: 21002
Epoch: [231]  [190/312]  eta: 0:01:08  lr: 0.000561  min_lr: 0.000561  loss: 3.9688 (3.8001)  weight_decay: 0.0500 (0.0500)  time: 0.5397  data: 0.1399  max mem: 21002
Epoch: [231]  [200/312]  eta: 0:01:02  lr: 0.000561  min_lr: 0.000561  loss: 3.9233 (3.7995)  weight_decay: 0.0500 (0.0500)  time: 0.4789  data: 0.1271  max mem: 21002
Epoch: [231]  [210/312]  eta: 0:00:57  lr: 0.000560  min_lr: 0.000560  loss: 3.7342 (3.7915)  weight_decay: 0.0500 (0.0500)  time: 0.6000  data: 0.1797  max mem: 21002
Epoch: [231]  [220/312]  eta: 0:00:51  lr: 0.000560  min_lr: 0.000560  loss: 3.7995 (3.7878)  weight_decay: 0.0500 (0.0500)  time: 0.5003  data: 0.0990  max mem: 21002
Epoch: [231]  [230/312]  eta: 0:00:46  lr: 0.000559  min_lr: 0.000559  loss: 3.8352 (3.7874)  weight_decay: 0.0500 (0.0500)  time: 0.5176  data: 0.0631  max mem: 21002
Epoch: [231]  [240/312]  eta: 0:00:40  lr: 0.000559  min_lr: 0.000559  loss: 3.8165 (3.7849)  weight_decay: 0.0500 (0.0500)  time: 0.6194  data: 0.0883  max mem: 21002
Epoch: [231]  [250/312]  eta: 0:00:34  lr: 0.000558  min_lr: 0.000558  loss: 3.9786 (3.7889)  weight_decay: 0.0500 (0.0500)  time: 0.5298  data: 0.1043  max mem: 21002
Epoch: [231]  [260/312]  eta: 0:00:29  lr: 0.000558  min_lr: 0.000558  loss: 4.0307 (3.7992)  weight_decay: 0.0500 (0.0500)  time: 0.5490  data: 0.1053  max mem: 21002
Epoch: [231]  [270/312]  eta: 0:00:23  lr: 0.000557  min_lr: 0.000557  loss: 3.9892 (3.7968)  weight_decay: 0.0500 (0.0500)  time: 0.5411  data: 0.0800  max mem: 21002
Epoch: [231]  [280/312]  eta: 0:00:17  lr: 0.000557  min_lr: 0.000557  loss: 3.8255 (3.7927)  weight_decay: 0.0500 (0.0500)  time: 0.5458  data: 0.0940  max mem: 21002
Epoch: [231]  [290/312]  eta: 0:00:12  lr: 0.000556  min_lr: 0.000556  loss: 3.8045 (3.7844)  weight_decay: 0.0500 (0.0500)  time: 0.5555  data: 0.1074  max mem: 21002
Epoch: [231]  [300/312]  eta: 0:00:06  lr: 0.000556  min_lr: 0.000556  loss: 3.7691 (3.7759)  weight_decay: 0.0500 (0.0500)  time: 0.4635  data: 0.0714  max mem: 21002
Epoch: [231]  [310/312]  eta: 0:00:01  lr: 0.000555  min_lr: 0.000555  loss: 3.7691 (3.7737)  weight_decay: 0.0500 (0.0500)  time: 0.3466  data: 0.0149  max mem: 21002
Epoch: [231]  [311/312]  eta: 0:00:00  lr: 0.000555  min_lr: 0.000555  loss: 3.8172 (3.7744)  weight_decay: 0.0500 (0.0500)  time: 0.3093  data: 0.0148  max mem: 21002
Epoch: [231] Total time: 0:02:50 (0.5470 s / it)
Averaged stats: lr: 0.000555  min_lr: 0.000555  loss: 3.8172 (3.6936)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.2347 (1.2347)  acc1: 77.8646 (77.8646)  acc5: 93.6198 (93.6198)  time: 8.8280  data: 8.7094  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.6704 (1.5536)  acc1: 68.4896 (69.9200)  acc5: 89.7135 (89.6960)  time: 1.0662  data: 0.9678  max mem: 21002
Test: Total time: 0:00:09 (1.0831 s / it)
* Acc@1 69.458 Acc@5 89.760 loss 1.552
Accuracy of the model on the 50000 test images: 69.5%
Max accuracy: 69.46%
Epoch: [232]  [  0/312]  eta: 1:02:15  lr: 0.000555  min_lr: 0.000555  loss: 4.0054 (4.0054)  weight_decay: 0.0500 (0.0500)  time: 11.9730  data: 8.8616  max mem: 21002
Epoch: [232]  [ 10/312]  eta: 0:08:02  lr: 0.000555  min_lr: 0.000555  loss: 3.8400 (3.6414)  weight_decay: 0.0500 (0.0500)  time: 1.5979  data: 0.9883  max mem: 21002
Epoch: [232]  [ 20/312]  eta: 0:05:11  lr: 0.000554  min_lr: 0.000554  loss: 3.7015 (3.5489)  weight_decay: 0.0500 (0.0500)  time: 0.5217  data: 0.1619  max mem: 21002
Epoch: [232]  [ 30/312]  eta: 0:03:50  lr: 0.000554  min_lr: 0.000554  loss: 3.2429 (3.4229)  weight_decay: 0.0500 (0.0500)  time: 0.3881  data: 0.0624  max mem: 21002
Epoch: [232]  [ 40/312]  eta: 0:03:13  lr: 0.000553  min_lr: 0.000553  loss: 3.3785 (3.4457)  weight_decay: 0.0500 (0.0500)  time: 0.3400  data: 0.0482  max mem: 21002
Epoch: [232]  [ 50/312]  eta: 0:03:00  lr: 0.000553  min_lr: 0.000553  loss: 3.7502 (3.4926)  weight_decay: 0.0500 (0.0500)  time: 0.4898  data: 0.1891  max mem: 21002
Epoch: [232]  [ 60/312]  eta: 0:02:39  lr: 0.000552  min_lr: 0.000552  loss: 3.8099 (3.5067)  weight_decay: 0.0500 (0.0500)  time: 0.4644  data: 0.1422  max mem: 21002
Epoch: [232]  [ 70/312]  eta: 0:02:37  lr: 0.000552  min_lr: 0.000552  loss: 3.7772 (3.5221)  weight_decay: 0.0500 (0.0500)  time: 0.5521  data: 0.1793  max mem: 21002
Epoch: [232]  [ 80/312]  eta: 0:02:27  lr: 0.000551  min_lr: 0.000551  loss: 3.7772 (3.5412)  weight_decay: 0.0500 (0.0500)  time: 0.6518  data: 0.2810  max mem: 21002
Epoch: [232]  [ 90/312]  eta: 0:02:15  lr: 0.000551  min_lr: 0.000551  loss: 3.6865 (3.5602)  weight_decay: 0.0500 (0.0500)  time: 0.4609  data: 0.1359  max mem: 21002
Epoch: [232]  [100/312]  eta: 0:02:11  lr: 0.000550  min_lr: 0.000550  loss: 3.8601 (3.5604)  weight_decay: 0.0500 (0.0500)  time: 0.5537  data: 0.1966  max mem: 21002
Epoch: [232]  [110/312]  eta: 0:01:59  lr: 0.000550  min_lr: 0.000550  loss: 3.2851 (3.5499)  weight_decay: 0.0500 (0.0500)  time: 0.5056  data: 0.1631  max mem: 21002
Epoch: [232]  [120/312]  eta: 0:01:54  lr: 0.000549  min_lr: 0.000549  loss: 3.5059 (3.5496)  weight_decay: 0.0500 (0.0500)  time: 0.4915  data: 0.1315  max mem: 21002
Epoch: [232]  [130/312]  eta: 0:01:49  lr: 0.000549  min_lr: 0.000549  loss: 3.6853 (3.5776)  weight_decay: 0.0500 (0.0500)  time: 0.6659  data: 0.2585  max mem: 21002
Epoch: [232]  [140/312]  eta: 0:01:40  lr: 0.000548  min_lr: 0.000548  loss: 3.8423 (3.5930)  weight_decay: 0.0500 (0.0500)  time: 0.4773  data: 0.1277  max mem: 21002
Epoch: [232]  [150/312]  eta: 0:01:35  lr: 0.000548  min_lr: 0.000548  loss: 3.7516 (3.5970)  weight_decay: 0.0500 (0.0500)  time: 0.5202  data: 0.1368  max mem: 21002
Epoch: [232]  [160/312]  eta: 0:01:29  lr: 0.000547  min_lr: 0.000547  loss: 3.3974 (3.5781)  weight_decay: 0.0500 (0.0500)  time: 0.6385  data: 0.2220  max mem: 21002
Epoch: [232]  [170/312]  eta: 0:01:21  lr: 0.000547  min_lr: 0.000547  loss: 3.6345 (3.5985)  weight_decay: 0.0500 (0.0500)  time: 0.4636  data: 0.1310  max mem: 21002
Epoch: [232]  [180/312]  eta: 0:01:16  lr: 0.000546  min_lr: 0.000546  loss: 3.9508 (3.6094)  weight_decay: 0.0500 (0.0500)  time: 0.4825  data: 0.1787  max mem: 21002
Epoch: [232]  [190/312]  eta: 0:01:08  lr: 0.000546  min_lr: 0.000546  loss: 3.8619 (3.6139)  weight_decay: 0.0500 (0.0500)  time: 0.4554  data: 0.1336  max mem: 21002
Epoch: [232]  [200/312]  eta: 0:01:03  lr: 0.000545  min_lr: 0.000545  loss: 3.7877 (3.6214)  weight_decay: 0.0500 (0.0500)  time: 0.5003  data: 0.1769  max mem: 21002
Epoch: [232]  [210/312]  eta: 0:00:58  lr: 0.000545  min_lr: 0.000545  loss: 3.8046 (3.6242)  weight_decay: 0.0500 (0.0500)  time: 0.6602  data: 0.3480  max mem: 21002
Epoch: [232]  [220/312]  eta: 0:00:51  lr: 0.000544  min_lr: 0.000544  loss: 3.8046 (3.6271)  weight_decay: 0.0500 (0.0500)  time: 0.5002  data: 0.1718  max mem: 21002
Epoch: [232]  [230/312]  eta: 0:00:46  lr: 0.000544  min_lr: 0.000544  loss: 3.7465 (3.6257)  weight_decay: 0.0500 (0.0500)  time: 0.5348  data: 0.1584  max mem: 21002
Epoch: [232]  [240/312]  eta: 0:00:40  lr: 0.000543  min_lr: 0.000543  loss: 3.7465 (3.6274)  weight_decay: 0.0500 (0.0500)  time: 0.5901  data: 0.2499  max mem: 21002
Epoch: [232]  [250/312]  eta: 0:00:34  lr: 0.000543  min_lr: 0.000543  loss: 3.9637 (3.6390)  weight_decay: 0.0500 (0.0500)  time: 0.4966  data: 0.1369  max mem: 21002
Epoch: [232]  [260/312]  eta: 0:00:29  lr: 0.000542  min_lr: 0.000542  loss: 3.9693 (3.6413)  weight_decay: 0.0500 (0.0500)  time: 0.5772  data: 0.1778  max mem: 21002
Epoch: [232]  [270/312]  eta: 0:00:23  lr: 0.000542  min_lr: 0.000542  loss: 3.6191 (3.6329)  weight_decay: 0.0500 (0.0500)  time: 0.4958  data: 0.1333  max mem: 21002
Epoch: [232]  [280/312]  eta: 0:00:18  lr: 0.000541  min_lr: 0.000541  loss: 3.6191 (3.6361)  weight_decay: 0.0500 (0.0500)  time: 0.5277  data: 0.0927  max mem: 21002
Epoch: [232]  [290/312]  eta: 0:00:12  lr: 0.000541  min_lr: 0.000541  loss: 3.7801 (3.6413)  weight_decay: 0.0500 (0.0500)  time: 0.5749  data: 0.1462  max mem: 21002
Epoch: [232]  [300/312]  eta: 0:00:06  lr: 0.000540  min_lr: 0.000540  loss: 3.7801 (3.6450)  weight_decay: 0.0500 (0.0500)  time: 0.4028  data: 0.0541  max mem: 21002
Epoch: [232]  [310/312]  eta: 0:00:01  lr: 0.000540  min_lr: 0.000540  loss: 3.5105 (3.6376)  weight_decay: 0.0500 (0.0500)  time: 0.3142  data: 0.0001  max mem: 21002
Epoch: [232]  [311/312]  eta: 0:00:00  lr: 0.000540  min_lr: 0.000540  loss: 3.5105 (3.6400)  weight_decay: 0.0500 (0.0500)  time: 0.2775  data: 0.0001  max mem: 21002
Epoch: [232] Total time: 0:02:50 (0.5466 s / it)
Averaged stats: lr: 0.000540  min_lr: 0.000540  loss: 3.5105 (3.6791)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 1.2230 (1.2230)  acc1: 76.5625 (76.5625)  acc5: 93.6198 (93.6198)  time: 8.9323  data: 8.8137  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.5813 (1.4709)  acc1: 68.4896 (69.9840)  acc5: 89.1927 (89.4400)  time: 1.0775  data: 0.9794  max mem: 21002
Test: Total time: 0:00:09 (1.1079 s / it)
* Acc@1 69.570 Acc@5 89.614 loss 1.477
Accuracy of the model on the 50000 test images: 69.6%
Max accuracy: 69.57%
Epoch: [233]  [  0/312]  eta: 1:01:06  lr: 0.000540  min_lr: 0.000540  loss: 3.4559 (3.4559)  weight_decay: 0.0500 (0.0500)  time: 11.7509  data: 10.5196  max mem: 21002
Epoch: [233]  [ 10/312]  eta: 0:08:09  lr: 0.000539  min_lr: 0.000539  loss: 3.9633 (3.7126)  weight_decay: 0.0500 (0.0500)  time: 1.6199  data: 1.0128  max mem: 21002
Epoch: [233]  [ 20/312]  eta: 0:05:10  lr: 0.000539  min_lr: 0.000539  loss: 3.9633 (3.7062)  weight_decay: 0.0500 (0.0500)  time: 0.5307  data: 0.1160  max mem: 21002
Epoch: [233]  [ 30/312]  eta: 0:03:50  lr: 0.000538  min_lr: 0.000538  loss: 3.9973 (3.7937)  weight_decay: 0.0500 (0.0500)  time: 0.3765  data: 0.0856  max mem: 21002
Epoch: [233]  [ 40/312]  eta: 0:03:20  lr: 0.000538  min_lr: 0.000538  loss: 3.9705 (3.7565)  weight_decay: 0.0500 (0.0500)  time: 0.3917  data: 0.0985  max mem: 21002
Epoch: [233]  [ 50/312]  eta: 0:03:06  lr: 0.000537  min_lr: 0.000537  loss: 3.6824 (3.7243)  weight_decay: 0.0500 (0.0500)  time: 0.5452  data: 0.2576  max mem: 21002
Epoch: [233]  [ 60/312]  eta: 0:02:41  lr: 0.000537  min_lr: 0.000537  loss: 3.7488 (3.7208)  weight_decay: 0.0500 (0.0500)  time: 0.4468  data: 0.1600  max mem: 21002
Epoch: [233]  [ 70/312]  eta: 0:02:36  lr: 0.000536  min_lr: 0.000536  loss: 3.7515 (3.7003)  weight_decay: 0.0500 (0.0500)  time: 0.4833  data: 0.1929  max mem: 21002
Epoch: [233]  [ 80/312]  eta: 0:02:30  lr: 0.000536  min_lr: 0.000536  loss: 3.5905 (3.6831)  weight_decay: 0.0500 (0.0500)  time: 0.6666  data: 0.3782  max mem: 21002
Epoch: [233]  [ 90/312]  eta: 0:02:15  lr: 0.000535  min_lr: 0.000535  loss: 3.5777 (3.6455)  weight_decay: 0.0500 (0.0500)  time: 0.4742  data: 0.1859  max mem: 21002
Epoch: [233]  [100/312]  eta: 0:02:10  lr: 0.000535  min_lr: 0.000535  loss: 3.7167 (3.6616)  weight_decay: 0.0500 (0.0500)  time: 0.4833  data: 0.1932  max mem: 21002
Epoch: [233]  [110/312]  eta: 0:01:58  lr: 0.000534  min_lr: 0.000534  loss: 3.8558 (3.6567)  weight_decay: 0.0500 (0.0500)  time: 0.4798  data: 0.1933  max mem: 21002
Epoch: [233]  [120/312]  eta: 0:01:53  lr: 0.000534  min_lr: 0.000534  loss: 3.8972 (3.6810)  weight_decay: 0.0500 (0.0500)  time: 0.4790  data: 0.1930  max mem: 21002
Epoch: [233]  [130/312]  eta: 0:01:49  lr: 0.000533  min_lr: 0.000533  loss: 3.9283 (3.6614)  weight_decay: 0.0500 (0.0500)  time: 0.6929  data: 0.3976  max mem: 21002
Epoch: [233]  [140/312]  eta: 0:01:39  lr: 0.000533  min_lr: 0.000533  loss: 3.8927 (3.6849)  weight_decay: 0.0500 (0.0500)  time: 0.5009  data: 0.2052  max mem: 21002
Epoch: [233]  [150/312]  eta: 0:01:35  lr: 0.000533  min_lr: 0.000533  loss: 3.8927 (3.6854)  weight_decay: 0.0500 (0.0500)  time: 0.4982  data: 0.1865  max mem: 21002
Epoch: [233]  [160/312]  eta: 0:01:29  lr: 0.000532  min_lr: 0.000532  loss: 3.6082 (3.6722)  weight_decay: 0.0500 (0.0500)  time: 0.6778  data: 0.3598  max mem: 21002
Epoch: [233]  [170/312]  eta: 0:01:21  lr: 0.000532  min_lr: 0.000532  loss: 3.6969 (3.6859)  weight_decay: 0.0500 (0.0500)  time: 0.4669  data: 0.1740  max mem: 21002
Epoch: [233]  [180/312]  eta: 0:01:16  lr: 0.000531  min_lr: 0.000531  loss: 3.8684 (3.6959)  weight_decay: 0.0500 (0.0500)  time: 0.5015  data: 0.1999  max mem: 21002
Epoch: [233]  [190/312]  eta: 0:01:09  lr: 0.000531  min_lr: 0.000531  loss: 3.8273 (3.6978)  weight_decay: 0.0500 (0.0500)  time: 0.5029  data: 0.2003  max mem: 21002
Epoch: [233]  [200/312]  eta: 0:01:04  lr: 0.000530  min_lr: 0.000530  loss: 3.7291 (3.6993)  weight_decay: 0.0500 (0.0500)  time: 0.4871  data: 0.1811  max mem: 21002
Epoch: [233]  [210/312]  eta: 0:00:58  lr: 0.000530  min_lr: 0.000530  loss: 3.8555 (3.7067)  weight_decay: 0.0500 (0.0500)  time: 0.6822  data: 0.3737  max mem: 21002
Epoch: [233]  [220/312]  eta: 0:00:51  lr: 0.000529  min_lr: 0.000529  loss: 3.8555 (3.7001)  weight_decay: 0.0500 (0.0500)  time: 0.4828  data: 0.1946  max mem: 21002
Epoch: [233]  [230/312]  eta: 0:00:46  lr: 0.000529  min_lr: 0.000529  loss: 3.8118 (3.6990)  weight_decay: 0.0500 (0.0500)  time: 0.5056  data: 0.2042  max mem: 21002
Epoch: [233]  [240/312]  eta: 0:00:41  lr: 0.000528  min_lr: 0.000528  loss: 3.8118 (3.6964)  weight_decay: 0.0500 (0.0500)  time: 0.7036  data: 0.3978  max mem: 21002
Epoch: [233]  [250/312]  eta: 0:00:35  lr: 0.000528  min_lr: 0.000528  loss: 3.8587 (3.7011)  weight_decay: 0.0500 (0.0500)  time: 0.5030  data: 0.1950  max mem: 21002
Epoch: [233]  [260/312]  eta: 0:00:29  lr: 0.000527  min_lr: 0.000527  loss: 3.8587 (3.7006)  weight_decay: 0.0500 (0.0500)  time: 0.4863  data: 0.1795  max mem: 21002
Epoch: [233]  [270/312]  eta: 0:00:23  lr: 0.000527  min_lr: 0.000527  loss: 3.8304 (3.7026)  weight_decay: 0.0500 (0.0500)  time: 0.5070  data: 0.1796  max mem: 21002
Epoch: [233]  [280/312]  eta: 0:00:18  lr: 0.000526  min_lr: 0.000526  loss: 3.8304 (3.7015)  weight_decay: 0.0500 (0.0500)  time: 0.4867  data: 0.1621  max mem: 21002
Epoch: [233]  [290/312]  eta: 0:00:12  lr: 0.000526  min_lr: 0.000526  loss: 3.6125 (3.7007)  weight_decay: 0.0500 (0.0500)  time: 0.5966  data: 0.2704  max mem: 21002
Epoch: [233]  [300/312]  eta: 0:00:06  lr: 0.000525  min_lr: 0.000525  loss: 3.9008 (3.7073)  weight_decay: 0.0500 (0.0500)  time: 0.4363  data: 0.1087  max mem: 21002
Epoch: [233]  [310/312]  eta: 0:00:01  lr: 0.000525  min_lr: 0.000525  loss: 3.9692 (3.7063)  weight_decay: 0.0500 (0.0500)  time: 0.2878  data: 0.0065  max mem: 21002
Epoch: [233]  [311/312]  eta: 0:00:00  lr: 0.000525  min_lr: 0.000525  loss: 3.9008 (3.7054)  weight_decay: 0.0500 (0.0500)  time: 0.2877  data: 0.0065  max mem: 21002
Epoch: [233] Total time: 0:02:51 (0.5489 s / it)
Averaged stats: lr: 0.000525  min_lr: 0.000525  loss: 3.9008 (3.6981)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.1907 (1.1907)  acc1: 77.9948 (77.9948)  acc5: 93.3594 (93.3594)  time: 8.3196  data: 8.2015  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.5492 (1.4786)  acc1: 69.2708 (69.8240)  acc5: 89.4531 (89.7440)  time: 1.0077  data: 0.9114  max mem: 21002
Test: Total time: 0:00:09 (1.0199 s / it)
* Acc@1 69.872 Acc@5 89.726 loss 1.480
Accuracy of the model on the 50000 test images: 69.9%
Max accuracy: 69.87%
Epoch: [234]  [  0/312]  eta: 0:59:06  lr: 0.000525  min_lr: 0.000525  loss: 4.1573 (4.1573)  weight_decay: 0.0500 (0.0500)  time: 11.3659  data: 8.6691  max mem: 21002
Epoch: [234]  [ 10/312]  eta: 0:07:54  lr: 0.000524  min_lr: 0.000524  loss: 3.8661 (3.8917)  weight_decay: 0.0500 (0.0500)  time: 1.5711  data: 0.9958  max mem: 21002
Epoch: [234]  [ 20/312]  eta: 0:05:17  lr: 0.000524  min_lr: 0.000524  loss: 3.8104 (3.6277)  weight_decay: 0.0500 (0.0500)  time: 0.5726  data: 0.1649  max mem: 21002
Epoch: [234]  [ 30/312]  eta: 0:03:54  lr: 0.000523  min_lr: 0.000523  loss: 3.6748 (3.6522)  weight_decay: 0.0500 (0.0500)  time: 0.4265  data: 0.0511  max mem: 21002
Epoch: [234]  [ 40/312]  eta: 0:03:30  lr: 0.000523  min_lr: 0.000523  loss: 3.7312 (3.5968)  weight_decay: 0.0500 (0.0500)  time: 0.4421  data: 0.0109  max mem: 21002
Epoch: [234]  [ 50/312]  eta: 0:03:15  lr: 0.000522  min_lr: 0.000522  loss: 3.5590 (3.5982)  weight_decay: 0.0500 (0.0500)  time: 0.6163  data: 0.0362  max mem: 21002
Epoch: [234]  [ 60/312]  eta: 0:02:51  lr: 0.000522  min_lr: 0.000522  loss: 3.4892 (3.5871)  weight_decay: 0.0500 (0.0500)  time: 0.4885  data: 0.0260  max mem: 21002
Epoch: [234]  [ 70/312]  eta: 0:02:42  lr: 0.000521  min_lr: 0.000521  loss: 3.9514 (3.6526)  weight_decay: 0.0500 (0.0500)  time: 0.4833  data: 0.0264  max mem: 21002
Epoch: [234]  [ 80/312]  eta: 0:02:33  lr: 0.000521  min_lr: 0.000521  loss: 3.9752 (3.6704)  weight_decay: 0.0500 (0.0500)  time: 0.6014  data: 0.0449  max mem: 21002
Epoch: [234]  [ 90/312]  eta: 0:02:20  lr: 0.000520  min_lr: 0.000520  loss: 3.8545 (3.6945)  weight_decay: 0.0500 (0.0500)  time: 0.4828  data: 0.0193  max mem: 21002
Epoch: [234]  [100/312]  eta: 0:02:15  lr: 0.000520  min_lr: 0.000520  loss: 3.8104 (3.6916)  weight_decay: 0.0500 (0.0500)  time: 0.5483  data: 0.0458  max mem: 21002
Epoch: [234]  [110/312]  eta: 0:02:02  lr: 0.000519  min_lr: 0.000519  loss: 3.5067 (3.6621)  weight_decay: 0.0500 (0.0500)  time: 0.4936  data: 0.0457  max mem: 21002
Epoch: [234]  [120/312]  eta: 0:01:56  lr: 0.000519  min_lr: 0.000519  loss: 3.3832 (3.6611)  weight_decay: 0.0500 (0.0500)  time: 0.4494  data: 0.0592  max mem: 21002
Epoch: [234]  [130/312]  eta: 0:01:50  lr: 0.000518  min_lr: 0.000518  loss: 3.7493 (3.6759)  weight_decay: 0.0500 (0.0500)  time: 0.6103  data: 0.1597  max mem: 21002
Epoch: [234]  [140/312]  eta: 0:01:41  lr: 0.000518  min_lr: 0.000518  loss: 3.7493 (3.6626)  weight_decay: 0.0500 (0.0500)  time: 0.4741  data: 0.1011  max mem: 21002
Epoch: [234]  [150/312]  eta: 0:01:36  lr: 0.000517  min_lr: 0.000517  loss: 3.4078 (3.6623)  weight_decay: 0.0500 (0.0500)  time: 0.5225  data: 0.0943  max mem: 21002
Epoch: [234]  [160/312]  eta: 0:01:30  lr: 0.000517  min_lr: 0.000517  loss: 3.8192 (3.6704)  weight_decay: 0.0500 (0.0500)  time: 0.6444  data: 0.1540  max mem: 21002
Epoch: [234]  [170/312]  eta: 0:01:23  lr: 0.000516  min_lr: 0.000516  loss: 3.8192 (3.6588)  weight_decay: 0.0500 (0.0500)  time: 0.5040  data: 0.0665  max mem: 21002
Epoch: [234]  [180/312]  eta: 0:01:18  lr: 0.000516  min_lr: 0.000516  loss: 3.5389 (3.6520)  weight_decay: 0.0500 (0.0500)  time: 0.5621  data: 0.0809  max mem: 21002
Epoch: [234]  [190/312]  eta: 0:01:10  lr: 0.000515  min_lr: 0.000515  loss: 3.5389 (3.6437)  weight_decay: 0.0500 (0.0500)  time: 0.4929  data: 0.0747  max mem: 21002
Epoch: [234]  [200/312]  eta: 0:01:05  lr: 0.000515  min_lr: 0.000515  loss: 3.6351 (3.6550)  weight_decay: 0.0500 (0.0500)  time: 0.4989  data: 0.0823  max mem: 21002
Epoch: [234]  [210/312]  eta: 0:00:59  lr: 0.000514  min_lr: 0.000514  loss: 3.9881 (3.6689)  weight_decay: 0.0500 (0.0500)  time: 0.6841  data: 0.1526  max mem: 21002
Epoch: [234]  [220/312]  eta: 0:00:52  lr: 0.000514  min_lr: 0.000514  loss: 3.8968 (3.6637)  weight_decay: 0.0500 (0.0500)  time: 0.4999  data: 0.0716  max mem: 21002
Epoch: [234]  [230/312]  eta: 0:00:47  lr: 0.000514  min_lr: 0.000514  loss: 3.7986 (3.6688)  weight_decay: 0.0500 (0.0500)  time: 0.5195  data: 0.0737  max mem: 21002
Epoch: [234]  [240/312]  eta: 0:00:41  lr: 0.000513  min_lr: 0.000513  loss: 3.7986 (3.6747)  weight_decay: 0.0500 (0.0500)  time: 0.5831  data: 0.0739  max mem: 21002
Epoch: [234]  [250/312]  eta: 0:00:35  lr: 0.000513  min_lr: 0.000513  loss: 3.7508 (3.6709)  weight_decay: 0.0500 (0.0500)  time: 0.4594  data: 0.0427  max mem: 21002
Epoch: [234]  [260/312]  eta: 0:00:29  lr: 0.000512  min_lr: 0.000512  loss: 3.8110 (3.6764)  weight_decay: 0.0500 (0.0500)  time: 0.5835  data: 0.0727  max mem: 21002
Epoch: [234]  [270/312]  eta: 0:00:23  lr: 0.000512  min_lr: 0.000512  loss: 3.7862 (3.6739)  weight_decay: 0.0500 (0.0500)  time: 0.5049  data: 0.0315  max mem: 21002
Epoch: [234]  [280/312]  eta: 0:00:18  lr: 0.000511  min_lr: 0.000511  loss: 3.8567 (3.6848)  weight_decay: 0.0500 (0.0500)  time: 0.4992  data: 0.0360  max mem: 21002
Epoch: [234]  [290/312]  eta: 0:00:12  lr: 0.000511  min_lr: 0.000511  loss: 3.9874 (3.6919)  weight_decay: 0.0500 (0.0500)  time: 0.5825  data: 0.0357  max mem: 21002
Epoch: [234]  [300/312]  eta: 0:00:06  lr: 0.000510  min_lr: 0.000510  loss: 3.9042 (3.6836)  weight_decay: 0.0500 (0.0500)  time: 0.3905  data: 0.0002  max mem: 21002
Epoch: [234]  [310/312]  eta: 0:00:01  lr: 0.000510  min_lr: 0.000510  loss: 3.3474 (3.6768)  weight_decay: 0.0500 (0.0500)  time: 0.2978  data: 0.0001  max mem: 21002
Epoch: [234]  [311/312]  eta: 0:00:00  lr: 0.000510  min_lr: 0.000510  loss: 3.4600 (3.6777)  weight_decay: 0.0500 (0.0500)  time: 0.2839  data: 0.0001  max mem: 21002
Epoch: [234] Total time: 0:02:52 (0.5525 s / it)
Averaged stats: lr: 0.000510  min_lr: 0.000510  loss: 3.4600 (3.6872)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.2347 (1.2347)  acc1: 76.1719 (76.1719)  acc5: 93.0990 (93.0990)  time: 8.4538  data: 8.3354  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.5250 (1.4604)  acc1: 69.0104 (69.4240)  acc5: 89.4531 (89.5520)  time: 1.0236  data: 0.9262  max mem: 21002
Test: Total time: 0:00:09 (1.0334 s / it)
* Acc@1 69.228 Acc@5 89.618 loss 1.465
Accuracy of the model on the 50000 test images: 69.2%
Max accuracy: 69.87%
Epoch: [235]  [  0/312]  eta: 1:08:13  lr: 0.000510  min_lr: 0.000510  loss: 3.5407 (3.5407)  weight_decay: 0.0500 (0.0500)  time: 13.1213  data: 10.8381  max mem: 21002
Epoch: [235]  [ 10/312]  eta: 0:09:25  lr: 0.000509  min_lr: 0.000509  loss: 3.8565 (3.6839)  weight_decay: 0.0500 (0.0500)  time: 1.8725  data: 0.9932  max mem: 21002
Epoch: [235]  [ 20/312]  eta: 0:05:35  lr: 0.000509  min_lr: 0.000509  loss: 3.8990 (3.7927)  weight_decay: 0.0500 (0.0500)  time: 0.5514  data: 0.0408  max mem: 21002
Epoch: [235]  [ 30/312]  eta: 0:04:05  lr: 0.000508  min_lr: 0.000508  loss: 4.0882 (3.8345)  weight_decay: 0.0500 (0.0500)  time: 0.3218  data: 0.0370  max mem: 21002
Epoch: [235]  [ 40/312]  eta: 0:03:19  lr: 0.000508  min_lr: 0.000508  loss: 3.6856 (3.7893)  weight_decay: 0.0500 (0.0500)  time: 0.2938  data: 0.0009  max mem: 21002
Epoch: [235]  [ 50/312]  eta: 0:02:53  lr: 0.000507  min_lr: 0.000507  loss: 3.3520 (3.7089)  weight_decay: 0.0500 (0.0500)  time: 0.3362  data: 0.0011  max mem: 21002
Epoch: [235]  [ 60/312]  eta: 0:02:42  lr: 0.000507  min_lr: 0.000507  loss: 3.6570 (3.7333)  weight_decay: 0.0500 (0.0500)  time: 0.4714  data: 0.0809  max mem: 21002
Epoch: [235]  [ 70/312]  eta: 0:02:34  lr: 0.000506  min_lr: 0.000506  loss: 3.8602 (3.7273)  weight_decay: 0.0500 (0.0500)  time: 0.5747  data: 0.1272  max mem: 21002
Epoch: [235]  [ 80/312]  eta: 0:02:20  lr: 0.000506  min_lr: 0.000506  loss: 3.8073 (3.7207)  weight_decay: 0.0500 (0.0500)  time: 0.4799  data: 0.0751  max mem: 21002
Epoch: [235]  [ 90/312]  eta: 0:02:15  lr: 0.000505  min_lr: 0.000505  loss: 3.4599 (3.6910)  weight_decay: 0.0500 (0.0500)  time: 0.5228  data: 0.0984  max mem: 21002
Epoch: [235]  [100/312]  eta: 0:02:09  lr: 0.000505  min_lr: 0.000505  loss: 3.8647 (3.6958)  weight_decay: 0.0500 (0.0500)  time: 0.6286  data: 0.0918  max mem: 21002
Epoch: [235]  [110/312]  eta: 0:02:00  lr: 0.000504  min_lr: 0.000504  loss: 3.8730 (3.7012)  weight_decay: 0.0500 (0.0500)  time: 0.5310  data: 0.0882  max mem: 21002
Epoch: [235]  [120/312]  eta: 0:01:56  lr: 0.000504  min_lr: 0.000504  loss: 3.8446 (3.7100)  weight_decay: 0.0500 (0.0500)  time: 0.5959  data: 0.1617  max mem: 21002
Epoch: [235]  [130/312]  eta: 0:01:46  lr: 0.000503  min_lr: 0.000503  loss: 3.6292 (3.6900)  weight_decay: 0.0500 (0.0500)  time: 0.5083  data: 0.0954  max mem: 21002
Epoch: [235]  [140/312]  eta: 0:01:41  lr: 0.000503  min_lr: 0.000503  loss: 3.7259 (3.6902)  weight_decay: 0.0500 (0.0500)  time: 0.4731  data: 0.0965  max mem: 21002
Epoch: [235]  [150/312]  eta: 0:01:35  lr: 0.000502  min_lr: 0.000502  loss: 3.8506 (3.7018)  weight_decay: 0.0500 (0.0500)  time: 0.6083  data: 0.1217  max mem: 21002
Epoch: [235]  [160/312]  eta: 0:01:27  lr: 0.000502  min_lr: 0.000502  loss: 3.8584 (3.7061)  weight_decay: 0.0500 (0.0500)  time: 0.5024  data: 0.1023  max mem: 21002
Epoch: [235]  [170/312]  eta: 0:01:22  lr: 0.000501  min_lr: 0.000501  loss: 3.8584 (3.6998)  weight_decay: 0.0500 (0.0500)  time: 0.5551  data: 0.1777  max mem: 21002
Epoch: [235]  [180/312]  eta: 0:01:16  lr: 0.000501  min_lr: 0.000501  loss: 3.6679 (3.6911)  weight_decay: 0.0500 (0.0500)  time: 0.6257  data: 0.1454  max mem: 21002
Epoch: [235]  [190/312]  eta: 0:01:10  lr: 0.000501  min_lr: 0.000501  loss: 3.5773 (3.6872)  weight_decay: 0.0500 (0.0500)  time: 0.5039  data: 0.1169  max mem: 21002
Epoch: [235]  [200/312]  eta: 0:01:05  lr: 0.000500  min_lr: 0.000500  loss: 3.8780 (3.6965)  weight_decay: 0.0500 (0.0500)  time: 0.5857  data: 0.1758  max mem: 21002
Epoch: [235]  [210/312]  eta: 0:00:58  lr: 0.000500  min_lr: 0.000500  loss: 3.7812 (3.6928)  weight_decay: 0.0500 (0.0500)  time: 0.5179  data: 0.1043  max mem: 21002
Epoch: [235]  [220/312]  eta: 0:00:52  lr: 0.000499  min_lr: 0.000499  loss: 3.7020 (3.6983)  weight_decay: 0.0500 (0.0500)  time: 0.4956  data: 0.0859  max mem: 21002
Epoch: [235]  [230/312]  eta: 0:00:47  lr: 0.000499  min_lr: 0.000499  loss: 3.6767 (3.7044)  weight_decay: 0.0500 (0.0500)  time: 0.6251  data: 0.0854  max mem: 21002
Epoch: [235]  [240/312]  eta: 0:00:41  lr: 0.000498  min_lr: 0.000498  loss: 3.8416 (3.7104)  weight_decay: 0.0500 (0.0500)  time: 0.5256  data: 0.1089  max mem: 21002
Epoch: [235]  [250/312]  eta: 0:00:35  lr: 0.000498  min_lr: 0.000498  loss: 3.8416 (3.7065)  weight_decay: 0.0500 (0.0500)  time: 0.5952  data: 0.2113  max mem: 21002
Epoch: [235]  [260/312]  eta: 0:00:29  lr: 0.000497  min_lr: 0.000497  loss: 3.8838 (3.7141)  weight_decay: 0.0500 (0.0500)  time: 0.5400  data: 0.1031  max mem: 21002
Epoch: [235]  [270/312]  eta: 0:00:23  lr: 0.000497  min_lr: 0.000497  loss: 3.8838 (3.7174)  weight_decay: 0.0500 (0.0500)  time: 0.4981  data: 0.0989  max mem: 21002
Epoch: [235]  [280/312]  eta: 0:00:18  lr: 0.000496  min_lr: 0.000496  loss: 3.8702 (3.7251)  weight_decay: 0.0500 (0.0500)  time: 0.6439  data: 0.2038  max mem: 21002
Epoch: [235]  [290/312]  eta: 0:00:12  lr: 0.000496  min_lr: 0.000496  loss: 3.8922 (3.7251)  weight_decay: 0.0500 (0.0500)  time: 0.4837  data: 0.1054  max mem: 21002
Epoch: [235]  [300/312]  eta: 0:00:06  lr: 0.000495  min_lr: 0.000495  loss: 3.8922 (3.7263)  weight_decay: 0.0500 (0.0500)  time: 0.4026  data: 0.0514  max mem: 21002
Epoch: [235]  [310/312]  eta: 0:00:01  lr: 0.000495  min_lr: 0.000495  loss: 3.9323 (3.7248)  weight_decay: 0.0500 (0.0500)  time: 0.3981  data: 0.0513  max mem: 21002
Epoch: [235]  [311/312]  eta: 0:00:00  lr: 0.000495  min_lr: 0.000495  loss: 3.5339 (3.7207)  weight_decay: 0.0500 (0.0500)  time: 0.3692  data: 0.0513  max mem: 21002
Epoch: [235] Total time: 0:02:53 (0.5565 s / it)
Averaged stats: lr: 0.000495  min_lr: 0.000495  loss: 3.5339 (3.6934)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.1387 (1.1387)  acc1: 77.7344 (77.7344)  acc5: 93.4896 (93.4896)  time: 8.7013  data: 8.5825  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4772 (1.4169)  acc1: 69.5312 (68.9760)  acc5: 90.1042 (89.6160)  time: 1.0791  data: 0.9818  max mem: 21002
Test: Total time: 0:00:09 (1.0894 s / it)
* Acc@1 69.022 Acc@5 89.470 loss 1.428
Accuracy of the model on the 50000 test images: 69.0%
Max accuracy: 69.87%
Epoch: [236]  [  0/312]  eta: 1:02:29  lr: 0.000495  min_lr: 0.000495  loss: 4.0137 (4.0137)  weight_decay: 0.0500 (0.0500)  time: 12.0164  data: 11.1782  max mem: 21002
Epoch: [236]  [ 10/312]  eta: 0:07:30  lr: 0.000494  min_lr: 0.000494  loss: 3.5156 (3.5803)  weight_decay: 0.0500 (0.0500)  time: 1.4913  data: 1.0639  max mem: 21002
Epoch: [236]  [ 20/312]  eta: 0:05:12  lr: 0.000494  min_lr: 0.000494  loss: 3.2590 (3.3710)  weight_decay: 0.0500 (0.0500)  time: 0.5234  data: 0.1428  max mem: 21002
Epoch: [236]  [ 30/312]  eta: 0:03:50  lr: 0.000493  min_lr: 0.000493  loss: 3.4744 (3.4930)  weight_decay: 0.0500 (0.0500)  time: 0.4485  data: 0.1169  max mem: 21002
Epoch: [236]  [ 40/312]  eta: 0:03:27  lr: 0.000493  min_lr: 0.000493  loss: 3.6846 (3.4869)  weight_decay: 0.0500 (0.0500)  time: 0.4369  data: 0.1027  max mem: 21002
Epoch: [236]  [ 50/312]  eta: 0:03:14  lr: 0.000492  min_lr: 0.000492  loss: 3.6272 (3.5283)  weight_decay: 0.0500 (0.0500)  time: 0.6201  data: 0.2145  max mem: 21002
Epoch: [236]  [ 60/312]  eta: 0:02:47  lr: 0.000492  min_lr: 0.000492  loss: 3.9257 (3.5373)  weight_decay: 0.0500 (0.0500)  time: 0.4717  data: 0.1125  max mem: 21002
Epoch: [236]  [ 70/312]  eta: 0:02:41  lr: 0.000491  min_lr: 0.000491  loss: 3.8635 (3.5331)  weight_decay: 0.0500 (0.0500)  time: 0.4746  data: 0.0957  max mem: 21002
Epoch: [236]  [ 80/312]  eta: 0:02:34  lr: 0.000491  min_lr: 0.000491  loss: 3.4227 (3.5230)  weight_decay: 0.0500 (0.0500)  time: 0.6577  data: 0.2207  max mem: 21002
Epoch: [236]  [ 90/312]  eta: 0:02:18  lr: 0.000490  min_lr: 0.000490  loss: 3.7801 (3.5644)  weight_decay: 0.0500 (0.0500)  time: 0.4749  data: 0.1257  max mem: 21002
Epoch: [236]  [100/312]  eta: 0:02:13  lr: 0.000490  min_lr: 0.000490  loss: 3.8727 (3.5585)  weight_decay: 0.0500 (0.0500)  time: 0.4970  data: 0.1338  max mem: 21002
Epoch: [236]  [110/312]  eta: 0:02:01  lr: 0.000490  min_lr: 0.000490  loss: 3.7087 (3.5742)  weight_decay: 0.0500 (0.0500)  time: 0.4944  data: 0.1346  max mem: 21002
Epoch: [236]  [120/312]  eta: 0:01:56  lr: 0.000489  min_lr: 0.000489  loss: 3.8332 (3.5753)  weight_decay: 0.0500 (0.0500)  time: 0.4716  data: 0.1215  max mem: 21002
Epoch: [236]  [130/312]  eta: 0:01:51  lr: 0.000489  min_lr: 0.000489  loss: 3.7880 (3.5825)  weight_decay: 0.0500 (0.0500)  time: 0.6722  data: 0.2364  max mem: 21002
Epoch: [236]  [140/312]  eta: 0:01:41  lr: 0.000488  min_lr: 0.000488  loss: 3.7880 (3.5841)  weight_decay: 0.0500 (0.0500)  time: 0.4896  data: 0.1171  max mem: 21002
Epoch: [236]  [150/312]  eta: 0:01:35  lr: 0.000488  min_lr: 0.000488  loss: 3.6056 (3.5832)  weight_decay: 0.0500 (0.0500)  time: 0.4613  data: 0.1048  max mem: 21002
Epoch: [236]  [160/312]  eta: 0:01:30  lr: 0.000487  min_lr: 0.000487  loss: 3.9413 (3.5929)  weight_decay: 0.0500 (0.0500)  time: 0.6339  data: 0.2436  max mem: 21002
Epoch: [236]  [170/312]  eta: 0:01:22  lr: 0.000487  min_lr: 0.000487  loss: 3.9762 (3.6087)  weight_decay: 0.0500 (0.0500)  time: 0.5173  data: 0.1401  max mem: 21002
Epoch: [236]  [180/312]  eta: 0:01:17  lr: 0.000486  min_lr: 0.000486  loss: 3.8252 (3.6046)  weight_decay: 0.0500 (0.0500)  time: 0.5238  data: 0.1032  max mem: 21002
Epoch: [236]  [190/312]  eta: 0:01:09  lr: 0.000486  min_lr: 0.000486  loss: 3.7174 (3.5981)  weight_decay: 0.0500 (0.0500)  time: 0.4848  data: 0.1032  max mem: 21002
Epoch: [236]  [200/312]  eta: 0:01:04  lr: 0.000485  min_lr: 0.000485  loss: 3.6670 (3.6001)  weight_decay: 0.0500 (0.0500)  time: 0.4665  data: 0.1129  max mem: 21002
Epoch: [236]  [210/312]  eta: 0:00:59  lr: 0.000485  min_lr: 0.000485  loss: 3.6670 (3.6021)  weight_decay: 0.0500 (0.0500)  time: 0.6488  data: 0.2415  max mem: 21002
Epoch: [236]  [220/312]  eta: 0:00:52  lr: 0.000484  min_lr: 0.000484  loss: 3.6043 (3.6046)  weight_decay: 0.0500 (0.0500)  time: 0.4889  data: 0.1293  max mem: 21002
Epoch: [236]  [230/312]  eta: 0:00:46  lr: 0.000484  min_lr: 0.000484  loss: 3.8842 (3.6144)  weight_decay: 0.0500 (0.0500)  time: 0.4901  data: 0.1022  max mem: 21002
Epoch: [236]  [240/312]  eta: 0:00:41  lr: 0.000483  min_lr: 0.000483  loss: 3.9006 (3.6255)  weight_decay: 0.0500 (0.0500)  time: 0.6808  data: 0.2122  max mem: 21002
Epoch: [236]  [250/312]  eta: 0:00:35  lr: 0.000483  min_lr: 0.000483  loss: 3.7357 (3.6278)  weight_decay: 0.0500 (0.0500)  time: 0.4795  data: 0.1106  max mem: 21002
Epoch: [236]  [260/312]  eta: 0:00:29  lr: 0.000482  min_lr: 0.000482  loss: 3.7582 (3.6263)  weight_decay: 0.0500 (0.0500)  time: 0.4837  data: 0.0995  max mem: 21002
Epoch: [236]  [270/312]  eta: 0:00:23  lr: 0.000482  min_lr: 0.000482  loss: 3.4497 (3.6148)  weight_decay: 0.0500 (0.0500)  time: 0.5800  data: 0.1003  max mem: 21002
Epoch: [236]  [280/312]  eta: 0:00:18  lr: 0.000482  min_lr: 0.000482  loss: 3.4497 (3.6204)  weight_decay: 0.0500 (0.0500)  time: 0.4883  data: 0.0808  max mem: 21002
Epoch: [236]  [290/312]  eta: 0:00:12  lr: 0.000481  min_lr: 0.000481  loss: 3.9968 (3.6293)  weight_decay: 0.0500 (0.0500)  time: 0.5267  data: 0.1254  max mem: 21002
Epoch: [236]  [300/312]  eta: 0:00:06  lr: 0.000481  min_lr: 0.000481  loss: 3.9198 (3.6338)  weight_decay: 0.0500 (0.0500)  time: 0.4176  data: 0.0458  max mem: 21002
Epoch: [236]  [310/312]  eta: 0:00:01  lr: 0.000480  min_lr: 0.000480  loss: 3.8630 (3.6333)  weight_decay: 0.0500 (0.0500)  time: 0.2835  data: 0.0001  max mem: 21002
Epoch: [236]  [311/312]  eta: 0:00:00  lr: 0.000480  min_lr: 0.000480  loss: 3.8630 (3.6332)  weight_decay: 0.0500 (0.0500)  time: 0.2835  data: 0.0001  max mem: 21002
Epoch: [236] Total time: 0:02:50 (0.5480 s / it)
Averaged stats: lr: 0.000480  min_lr: 0.000480  loss: 3.8630 (3.6870)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:10  loss: 1.2194 (1.2194)  acc1: 75.5208 (75.5208)  acc5: 93.7500 (93.7500)  time: 7.7846  data: 7.6662  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.6184 (1.4793)  acc1: 68.4896 (70.0000)  acc5: 88.9323 (89.6480)  time: 1.0187  data: 0.9224  max mem: 21002
Test: Total time: 0:00:09 (1.0360 s / it)
* Acc@1 69.628 Acc@5 89.790 loss 1.482
Accuracy of the model on the 50000 test images: 69.6%
Max accuracy: 69.87%
Epoch: [237]  [  0/312]  eta: 1:03:28  lr: 0.000480  min_lr: 0.000480  loss: 3.5954 (3.5954)  weight_decay: 0.0500 (0.0500)  time: 12.2063  data: 11.9151  max mem: 21002
Epoch: [237]  [ 10/312]  eta: 0:07:33  lr: 0.000480  min_lr: 0.000480  loss: 3.7325 (3.6144)  weight_decay: 0.0500 (0.0500)  time: 1.5033  data: 1.0839  max mem: 21002
Epoch: [237]  [ 20/312]  eta: 0:05:04  lr: 0.000479  min_lr: 0.000479  loss: 3.7148 (3.5910)  weight_decay: 0.0500 (0.0500)  time: 0.4853  data: 0.1040  max mem: 21002
Epoch: [237]  [ 30/312]  eta: 0:03:45  lr: 0.000479  min_lr: 0.000479  loss: 3.3782 (3.5671)  weight_decay: 0.0500 (0.0500)  time: 0.4109  data: 0.1040  max mem: 21002
Epoch: [237]  [ 40/312]  eta: 0:03:20  lr: 0.000478  min_lr: 0.000478  loss: 3.6979 (3.6280)  weight_decay: 0.0500 (0.0500)  time: 0.4168  data: 0.0845  max mem: 21002
Epoch: [237]  [ 50/312]  eta: 0:03:19  lr: 0.000478  min_lr: 0.000478  loss: 3.6982 (3.5952)  weight_decay: 0.0500 (0.0500)  time: 0.7011  data: 0.2399  max mem: 21002
Epoch: [237]  [ 60/312]  eta: 0:02:52  lr: 0.000477  min_lr: 0.000477  loss: 3.7059 (3.6180)  weight_decay: 0.0500 (0.0500)  time: 0.5698  data: 0.1564  max mem: 21002
Epoch: [237]  [ 70/312]  eta: 0:02:42  lr: 0.000477  min_lr: 0.000477  loss: 3.6055 (3.6113)  weight_decay: 0.0500 (0.0500)  time: 0.4510  data: 0.0628  max mem: 21002
Epoch: [237]  [ 80/312]  eta: 0:02:31  lr: 0.000476  min_lr: 0.000476  loss: 3.5028 (3.5945)  weight_decay: 0.0500 (0.0500)  time: 0.5648  data: 0.1303  max mem: 21002
Epoch: [237]  [ 90/312]  eta: 0:02:20  lr: 0.000476  min_lr: 0.000476  loss: 3.7250 (3.6175)  weight_decay: 0.0500 (0.0500)  time: 0.4827  data: 0.1020  max mem: 21002
Epoch: [237]  [100/312]  eta: 0:02:12  lr: 0.000475  min_lr: 0.000475  loss: 3.7315 (3.6232)  weight_decay: 0.0500 (0.0500)  time: 0.5088  data: 0.1401  max mem: 21002
Epoch: [237]  [110/312]  eta: 0:02:00  lr: 0.000475  min_lr: 0.000475  loss: 3.7032 (3.6218)  weight_decay: 0.0500 (0.0500)  time: 0.4277  data: 0.1068  max mem: 21002
Epoch: [237]  [120/312]  eta: 0:01:56  lr: 0.000474  min_lr: 0.000474  loss: 3.7131 (3.6144)  weight_decay: 0.0500 (0.0500)  time: 0.5162  data: 0.1842  max mem: 21002
Epoch: [237]  [130/312]  eta: 0:01:51  lr: 0.000474  min_lr: 0.000474  loss: 3.4206 (3.5965)  weight_decay: 0.0500 (0.0500)  time: 0.7276  data: 0.3346  max mem: 21002
Epoch: [237]  [140/312]  eta: 0:01:41  lr: 0.000474  min_lr: 0.000474  loss: 3.7550 (3.6036)  weight_decay: 0.0500 (0.0500)  time: 0.4993  data: 0.1511  max mem: 21002
Epoch: [237]  [150/312]  eta: 0:01:37  lr: 0.000473  min_lr: 0.000473  loss: 3.8395 (3.6054)  weight_decay: 0.0500 (0.0500)  time: 0.5322  data: 0.1142  max mem: 21002
Epoch: [237]  [160/312]  eta: 0:01:30  lr: 0.000473  min_lr: 0.000473  loss: 3.7198 (3.5999)  weight_decay: 0.0500 (0.0500)  time: 0.6100  data: 0.1199  max mem: 21002
Epoch: [237]  [170/312]  eta: 0:01:23  lr: 0.000472  min_lr: 0.000472  loss: 3.7582 (3.6109)  weight_decay: 0.0500 (0.0500)  time: 0.4709  data: 0.0802  max mem: 21002
Epoch: [237]  [180/312]  eta: 0:01:18  lr: 0.000472  min_lr: 0.000472  loss: 3.7582 (3.6109)  weight_decay: 0.0500 (0.0500)  time: 0.5696  data: 0.1472  max mem: 21002
Epoch: [237]  [190/312]  eta: 0:01:10  lr: 0.000471  min_lr: 0.000471  loss: 3.8017 (3.6248)  weight_decay: 0.0500 (0.0500)  time: 0.4684  data: 0.0733  max mem: 21002
Epoch: [237]  [200/312]  eta: 0:01:04  lr: 0.000471  min_lr: 0.000471  loss: 3.5994 (3.6188)  weight_decay: 0.0500 (0.0500)  time: 0.4717  data: 0.0728  max mem: 21002
Epoch: [237]  [210/312]  eta: 0:00:59  lr: 0.000470  min_lr: 0.000470  loss: 3.5119 (3.6201)  weight_decay: 0.0500 (0.0500)  time: 0.7021  data: 0.1898  max mem: 21002
Epoch: [237]  [220/312]  eta: 0:00:52  lr: 0.000470  min_lr: 0.000470  loss: 3.8944 (3.6440)  weight_decay: 0.0500 (0.0500)  time: 0.5272  data: 0.1177  max mem: 21002
Epoch: [237]  [230/312]  eta: 0:00:47  lr: 0.000469  min_lr: 0.000469  loss: 3.7047 (3.6330)  weight_decay: 0.0500 (0.0500)  time: 0.4368  data: 0.0756  max mem: 21002
Epoch: [237]  [240/312]  eta: 0:00:41  lr: 0.000469  min_lr: 0.000469  loss: 3.5526 (3.6302)  weight_decay: 0.0500 (0.0500)  time: 0.5309  data: 0.1086  max mem: 21002
Epoch: [237]  [250/312]  eta: 0:00:35  lr: 0.000468  min_lr: 0.000468  loss: 3.8160 (3.6357)  weight_decay: 0.0500 (0.0500)  time: 0.5066  data: 0.1111  max mem: 21002
Epoch: [237]  [260/312]  eta: 0:00:29  lr: 0.000468  min_lr: 0.000468  loss: 3.6440 (3.6370)  weight_decay: 0.0500 (0.0500)  time: 0.5542  data: 0.1717  max mem: 21002
Epoch: [237]  [270/312]  eta: 0:00:23  lr: 0.000468  min_lr: 0.000468  loss: 3.5864 (3.6372)  weight_decay: 0.0500 (0.0500)  time: 0.4383  data: 0.0943  max mem: 21002
Epoch: [237]  [280/312]  eta: 0:00:18  lr: 0.000467  min_lr: 0.000467  loss: 3.9122 (3.6472)  weight_decay: 0.0500 (0.0500)  time: 0.4794  data: 0.1406  max mem: 21002
Epoch: [237]  [290/312]  eta: 0:00:12  lr: 0.000467  min_lr: 0.000467  loss: 3.9830 (3.6558)  weight_decay: 0.0500 (0.0500)  time: 0.6031  data: 0.2171  max mem: 21002
Epoch: [237]  [300/312]  eta: 0:00:06  lr: 0.000466  min_lr: 0.000466  loss: 3.9351 (3.6517)  weight_decay: 0.0500 (0.0500)  time: 0.4109  data: 0.0770  max mem: 21002
Epoch: [237]  [310/312]  eta: 0:00:01  lr: 0.000466  min_lr: 0.000466  loss: 3.5669 (3.6479)  weight_decay: 0.0500 (0.0500)  time: 0.2773  data: 0.0001  max mem: 21002
Epoch: [237]  [311/312]  eta: 0:00:00  lr: 0.000466  min_lr: 0.000466  loss: 3.3909 (3.6445)  weight_decay: 0.0500 (0.0500)  time: 0.2773  data: 0.0001  max mem: 21002
Epoch: [237] Total time: 0:02:50 (0.5470 s / it)
Averaged stats: lr: 0.000466  min_lr: 0.000466  loss: 3.3909 (3.6701)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:11  loss: 1.0949 (1.0949)  acc1: 78.7760 (78.7760)  acc5: 94.7917 (94.7917)  time: 7.9596  data: 7.8412  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4766 (1.4073)  acc1: 70.7031 (70.9920)  acc5: 89.3229 (90.0640)  time: 1.0061  data: 0.9083  max mem: 21002
Test: Total time: 0:00:09 (1.0161 s / it)
* Acc@1 70.618 Acc@5 90.278 loss 1.417
Accuracy of the model on the 50000 test images: 70.6%
Max accuracy: 70.62%
Epoch: [238]  [  0/312]  eta: 1:01:25  lr: 0.000466  min_lr: 0.000466  loss: 3.8881 (3.8881)  weight_decay: 0.0500 (0.0500)  time: 11.8128  data: 8.5155  max mem: 21002
Epoch: [238]  [ 10/312]  eta: 0:08:02  lr: 0.000465  min_lr: 0.000465  loss: 3.8752 (3.6229)  weight_decay: 0.0500 (0.0500)  time: 1.5968  data: 1.0396  max mem: 21002
Epoch: [238]  [ 20/312]  eta: 0:05:19  lr: 0.000465  min_lr: 0.000465  loss: 3.6480 (3.4993)  weight_decay: 0.0500 (0.0500)  time: 0.5568  data: 0.2016  max mem: 21002
Epoch: [238]  [ 30/312]  eta: 0:03:56  lr: 0.000464  min_lr: 0.000464  loss: 3.6370 (3.5521)  weight_decay: 0.0500 (0.0500)  time: 0.4200  data: 0.0559  max mem: 21002
Epoch: [238]  [ 40/312]  eta: 0:03:18  lr: 0.000464  min_lr: 0.000464  loss: 3.6370 (3.5866)  weight_decay: 0.0500 (0.0500)  time: 0.3513  data: 0.0314  max mem: 21002
Epoch: [238]  [ 50/312]  eta: 0:03:12  lr: 0.000463  min_lr: 0.000463  loss: 3.9650 (3.6627)  weight_decay: 0.0500 (0.0500)  time: 0.5770  data: 0.1698  max mem: 21002
Epoch: [238]  [ 60/312]  eta: 0:02:46  lr: 0.000463  min_lr: 0.000463  loss: 3.9650 (3.6300)  weight_decay: 0.0500 (0.0500)  time: 0.5193  data: 0.1391  max mem: 21002
Epoch: [238]  [ 70/312]  eta: 0:02:37  lr: 0.000462  min_lr: 0.000462  loss: 3.5338 (3.6360)  weight_decay: 0.0500 (0.0500)  time: 0.4400  data: 0.0903  max mem: 21002
Epoch: [238]  [ 80/312]  eta: 0:02:27  lr: 0.000462  min_lr: 0.000462  loss: 3.7994 (3.6362)  weight_decay: 0.0500 (0.0500)  time: 0.5565  data: 0.1723  max mem: 21002
Epoch: [238]  [ 90/312]  eta: 0:02:15  lr: 0.000461  min_lr: 0.000461  loss: 3.8123 (3.6389)  weight_decay: 0.0500 (0.0500)  time: 0.4711  data: 0.1086  max mem: 21002
Epoch: [238]  [100/312]  eta: 0:02:12  lr: 0.000461  min_lr: 0.000461  loss: 3.8961 (3.6427)  weight_decay: 0.0500 (0.0500)  time: 0.5799  data: 0.1574  max mem: 21002
Epoch: [238]  [110/312]  eta: 0:02:00  lr: 0.000461  min_lr: 0.000461  loss: 3.9180 (3.6539)  weight_decay: 0.0500 (0.0500)  time: 0.5116  data: 0.1315  max mem: 21002
Epoch: [238]  [120/312]  eta: 0:01:55  lr: 0.000460  min_lr: 0.000460  loss: 3.9360 (3.6469)  weight_decay: 0.0500 (0.0500)  time: 0.4867  data: 0.1186  max mem: 21002
Epoch: [238]  [130/312]  eta: 0:01:49  lr: 0.000460  min_lr: 0.000460  loss: 3.8139 (3.6458)  weight_decay: 0.0500 (0.0500)  time: 0.6425  data: 0.1890  max mem: 21002
Epoch: [238]  [140/312]  eta: 0:01:40  lr: 0.000459  min_lr: 0.000459  loss: 3.7412 (3.6415)  weight_decay: 0.0500 (0.0500)  time: 0.4818  data: 0.1109  max mem: 21002
Epoch: [238]  [150/312]  eta: 0:01:35  lr: 0.000459  min_lr: 0.000459  loss: 3.6195 (3.6477)  weight_decay: 0.0500 (0.0500)  time: 0.5131  data: 0.1393  max mem: 21002
Epoch: [238]  [160/312]  eta: 0:01:28  lr: 0.000458  min_lr: 0.000458  loss: 3.7962 (3.6647)  weight_decay: 0.0500 (0.0500)  time: 0.5721  data: 0.1318  max mem: 21002
Epoch: [238]  [170/312]  eta: 0:01:21  lr: 0.000458  min_lr: 0.000458  loss: 3.9569 (3.6834)  weight_decay: 0.0500 (0.0500)  time: 0.4797  data: 0.1132  max mem: 21002
Epoch: [238]  [180/312]  eta: 0:01:17  lr: 0.000457  min_lr: 0.000457  loss: 3.8840 (3.6829)  weight_decay: 0.0500 (0.0500)  time: 0.5883  data: 0.1787  max mem: 21002
Epoch: [238]  [190/312]  eta: 0:01:09  lr: 0.000457  min_lr: 0.000457  loss: 3.6264 (3.6664)  weight_decay: 0.0500 (0.0500)  time: 0.4963  data: 0.0984  max mem: 21002
Epoch: [238]  [200/312]  eta: 0:01:04  lr: 0.000456  min_lr: 0.000456  loss: 3.5871 (3.6624)  weight_decay: 0.0500 (0.0500)  time: 0.4858  data: 0.0826  max mem: 21002
Epoch: [238]  [210/312]  eta: 0:00:58  lr: 0.000456  min_lr: 0.000456  loss: 3.8203 (3.6729)  weight_decay: 0.0500 (0.0500)  time: 0.6525  data: 0.1379  max mem: 21002
Epoch: [238]  [220/312]  eta: 0:00:52  lr: 0.000456  min_lr: 0.000456  loss: 3.8991 (3.6813)  weight_decay: 0.0500 (0.0500)  time: 0.4959  data: 0.0942  max mem: 21002
Epoch: [238]  [230/312]  eta: 0:00:46  lr: 0.000455  min_lr: 0.000455  loss: 3.8991 (3.6894)  weight_decay: 0.0500 (0.0500)  time: 0.4828  data: 0.0976  max mem: 21002
Epoch: [238]  [240/312]  eta: 0:00:40  lr: 0.000455  min_lr: 0.000455  loss: 3.9879 (3.7048)  weight_decay: 0.0500 (0.0500)  time: 0.5583  data: 0.0660  max mem: 21002
Epoch: [238]  [250/312]  eta: 0:00:34  lr: 0.000454  min_lr: 0.000454  loss: 3.8059 (3.6945)  weight_decay: 0.0500 (0.0500)  time: 0.4836  data: 0.0892  max mem: 21002
Epoch: [238]  [260/312]  eta: 0:00:29  lr: 0.000454  min_lr: 0.000454  loss: 3.2783 (3.6801)  weight_decay: 0.0500 (0.0500)  time: 0.5968  data: 0.1730  max mem: 21002
Epoch: [238]  [270/312]  eta: 0:00:23  lr: 0.000453  min_lr: 0.000453  loss: 3.5312 (3.6805)  weight_decay: 0.0500 (0.0500)  time: 0.5150  data: 0.0910  max mem: 21002
Epoch: [238]  [280/312]  eta: 0:00:18  lr: 0.000453  min_lr: 0.000453  loss: 3.7457 (3.6678)  weight_decay: 0.0500 (0.0500)  time: 0.4862  data: 0.0878  max mem: 21002
Epoch: [238]  [290/312]  eta: 0:00:12  lr: 0.000452  min_lr: 0.000452  loss: 3.5846 (3.6715)  weight_decay: 0.0500 (0.0500)  time: 0.6427  data: 0.1517  max mem: 21002
Epoch: [238]  [300/312]  eta: 0:00:06  lr: 0.000452  min_lr: 0.000452  loss: 3.9338 (3.6754)  weight_decay: 0.0500 (0.0500)  time: 0.4401  data: 0.0643  max mem: 21002
Epoch: [238]  [310/312]  eta: 0:00:01  lr: 0.000451  min_lr: 0.000451  loss: 3.8092 (3.6791)  weight_decay: 0.0500 (0.0500)  time: 0.2785  data: 0.0001  max mem: 21002
Epoch: [238]  [311/312]  eta: 0:00:00  lr: 0.000451  min_lr: 0.000451  loss: 3.8092 (3.6815)  weight_decay: 0.0500 (0.0500)  time: 0.2784  data: 0.0001  max mem: 21002
Epoch: [238] Total time: 0:02:51 (0.5483 s / it)
Averaged stats: lr: 0.000451  min_lr: 0.000451  loss: 3.8092 (3.6847)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.2790 (1.2790)  acc1: 79.2969 (79.2969)  acc5: 94.0104 (94.0104)  time: 8.7994  data: 8.6824  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.5342 (1.4851)  acc1: 69.2708 (70.9600)  acc5: 89.4531 (90.4000)  time: 1.0669  data: 0.9648  max mem: 21002
Test: Total time: 0:00:09 (1.1000 s / it)
* Acc@1 70.642 Acc@5 90.144 loss 1.499
Accuracy of the model on the 50000 test images: 70.6%
Max accuracy: 70.64%
Epoch: [239]  [  0/312]  eta: 0:59:53  lr: 0.000451  min_lr: 0.000451  loss: 4.2469 (4.2469)  weight_decay: 0.0500 (0.0500)  time: 11.5183  data: 10.5436  max mem: 21002
Epoch: [239]  [ 10/312]  eta: 0:08:10  lr: 0.000451  min_lr: 0.000451  loss: 3.4868 (3.4234)  weight_decay: 0.0500 (0.0500)  time: 1.6240  data: 1.0276  max mem: 21002
Epoch: [239]  [ 20/312]  eta: 0:04:55  lr: 0.000450  min_lr: 0.000450  loss: 3.6476 (3.6415)  weight_decay: 0.0500 (0.0500)  time: 0.4856  data: 0.0383  max mem: 21002
Epoch: [239]  [ 30/312]  eta: 0:03:40  lr: 0.000450  min_lr: 0.000450  loss: 3.8633 (3.6414)  weight_decay: 0.0500 (0.0500)  time: 0.3180  data: 0.0049  max mem: 21002
Epoch: [239]  [ 40/312]  eta: 0:03:06  lr: 0.000449  min_lr: 0.000449  loss: 3.7707 (3.6340)  weight_decay: 0.0500 (0.0500)  time: 0.3445  data: 0.0525  max mem: 21002
Epoch: [239]  [ 50/312]  eta: 0:02:57  lr: 0.000449  min_lr: 0.000449  loss: 3.6091 (3.6356)  weight_decay: 0.0500 (0.0500)  time: 0.5212  data: 0.2127  max mem: 21002
Epoch: [239]  [ 60/312]  eta: 0:02:35  lr: 0.000449  min_lr: 0.000449  loss: 3.5478 (3.5923)  weight_decay: 0.0500 (0.0500)  time: 0.4718  data: 0.1651  max mem: 21002
Epoch: [239]  [ 70/312]  eta: 0:02:32  lr: 0.000448  min_lr: 0.000448  loss: 3.8460 (3.6355)  weight_decay: 0.0500 (0.0500)  time: 0.4999  data: 0.1808  max mem: 21002
Epoch: [239]  [ 80/312]  eta: 0:02:23  lr: 0.000448  min_lr: 0.000448  loss: 3.8460 (3.6197)  weight_decay: 0.0500 (0.0500)  time: 0.6318  data: 0.2712  max mem: 21002
Epoch: [239]  [ 90/312]  eta: 0:02:11  lr: 0.000447  min_lr: 0.000447  loss: 3.4965 (3.5999)  weight_decay: 0.0500 (0.0500)  time: 0.4648  data: 0.1355  max mem: 21002
Epoch: [239]  [100/312]  eta: 0:02:06  lr: 0.000447  min_lr: 0.000447  loss: 3.7602 (3.6309)  weight_decay: 0.0500 (0.0500)  time: 0.5079  data: 0.1828  max mem: 21002
Epoch: [239]  [110/312]  eta: 0:01:55  lr: 0.000446  min_lr: 0.000446  loss: 3.8868 (3.6408)  weight_decay: 0.0500 (0.0500)  time: 0.4771  data: 0.1539  max mem: 21002
Epoch: [239]  [120/312]  eta: 0:01:53  lr: 0.000446  min_lr: 0.000446  loss: 3.7419 (3.6308)  weight_decay: 0.0500 (0.0500)  time: 0.5445  data: 0.1705  max mem: 21002
Epoch: [239]  [130/312]  eta: 0:01:47  lr: 0.000445  min_lr: 0.000445  loss: 3.7854 (3.6402)  weight_decay: 0.0500 (0.0500)  time: 0.6816  data: 0.2134  max mem: 21002
Epoch: [239]  [140/312]  eta: 0:01:38  lr: 0.000445  min_lr: 0.000445  loss: 3.6709 (3.6249)  weight_decay: 0.0500 (0.0500)  time: 0.4652  data: 0.0873  max mem: 21002
Epoch: [239]  [150/312]  eta: 0:01:33  lr: 0.000445  min_lr: 0.000445  loss: 3.6455 (3.6338)  weight_decay: 0.0500 (0.0500)  time: 0.5143  data: 0.1415  max mem: 21002
Epoch: [239]  [160/312]  eta: 0:01:27  lr: 0.000444  min_lr: 0.000444  loss: 3.8351 (3.6363)  weight_decay: 0.0500 (0.0500)  time: 0.5951  data: 0.1358  max mem: 21002
Epoch: [239]  [170/312]  eta: 0:01:20  lr: 0.000444  min_lr: 0.000444  loss: 3.7602 (3.6306)  weight_decay: 0.0500 (0.0500)  time: 0.4712  data: 0.1031  max mem: 21002
Epoch: [239]  [180/312]  eta: 0:01:15  lr: 0.000443  min_lr: 0.000443  loss: 3.3608 (3.6185)  weight_decay: 0.0500 (0.0500)  time: 0.5343  data: 0.1480  max mem: 21002
Epoch: [239]  [190/312]  eta: 0:01:08  lr: 0.000443  min_lr: 0.000443  loss: 3.3608 (3.6075)  weight_decay: 0.0500 (0.0500)  time: 0.5035  data: 0.1143  max mem: 21002
Epoch: [239]  [200/312]  eta: 0:01:03  lr: 0.000442  min_lr: 0.000442  loss: 3.6802 (3.6090)  weight_decay: 0.0500 (0.0500)  time: 0.5819  data: 0.1715  max mem: 21002
Epoch: [239]  [210/312]  eta: 0:00:57  lr: 0.000442  min_lr: 0.000442  loss: 3.6802 (3.6059)  weight_decay: 0.0500 (0.0500)  time: 0.6250  data: 0.1255  max mem: 21002
Epoch: [239]  [220/312]  eta: 0:00:51  lr: 0.000441  min_lr: 0.000441  loss: 3.5446 (3.6069)  weight_decay: 0.0500 (0.0500)  time: 0.4277  data: 0.0545  max mem: 21002
Epoch: [239]  [230/312]  eta: 0:00:46  lr: 0.000441  min_lr: 0.000441  loss: 3.7586 (3.6076)  weight_decay: 0.0500 (0.0500)  time: 0.5823  data: 0.1759  max mem: 21002
Epoch: [239]  [240/312]  eta: 0:00:40  lr: 0.000440  min_lr: 0.000440  loss: 3.8105 (3.6207)  weight_decay: 0.0500 (0.0500)  time: 0.5938  data: 0.1229  max mem: 21002
Epoch: [239]  [250/312]  eta: 0:00:34  lr: 0.000440  min_lr: 0.000440  loss: 3.8342 (3.6209)  weight_decay: 0.0500 (0.0500)  time: 0.4552  data: 0.0686  max mem: 21002
Epoch: [239]  [260/312]  eta: 0:00:29  lr: 0.000440  min_lr: 0.000440  loss: 3.8318 (3.6259)  weight_decay: 0.0500 (0.0500)  time: 0.5293  data: 0.0924  max mem: 21002
Epoch: [239]  [270/312]  eta: 0:00:23  lr: 0.000439  min_lr: 0.000439  loss: 3.9188 (3.6350)  weight_decay: 0.0500 (0.0500)  time: 0.4932  data: 0.0900  max mem: 21002
Epoch: [239]  [280/312]  eta: 0:00:17  lr: 0.000439  min_lr: 0.000439  loss: 3.6966 (3.6292)  weight_decay: 0.0500 (0.0500)  time: 0.5551  data: 0.1625  max mem: 21002
Epoch: [239]  [290/312]  eta: 0:00:12  lr: 0.000438  min_lr: 0.000438  loss: 3.6966 (3.6338)  weight_decay: 0.0500 (0.0500)  time: 0.5835  data: 0.0972  max mem: 21002
Epoch: [239]  [300/312]  eta: 0:00:06  lr: 0.000438  min_lr: 0.000438  loss: 3.9145 (3.6387)  weight_decay: 0.0500 (0.0500)  time: 0.4394  data: 0.0621  max mem: 21002
Epoch: [239]  [310/312]  eta: 0:00:01  lr: 0.000437  min_lr: 0.000437  loss: 3.8827 (3.6453)  weight_decay: 0.0500 (0.0500)  time: 0.3382  data: 0.0621  max mem: 21002
Epoch: [239]  [311/312]  eta: 0:00:00  lr: 0.000437  min_lr: 0.000437  loss: 3.9145 (3.6478)  weight_decay: 0.0500 (0.0500)  time: 0.3245  data: 0.0478  max mem: 21002
Epoch: [239] Total time: 0:02:49 (0.5442 s / it)
Averaged stats: lr: 0.000437  min_lr: 0.000437  loss: 3.9145 (3.6634)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.2604 (1.2604)  acc1: 79.6875 (79.6875)  acc5: 94.0104 (94.0104)  time: 8.4901  data: 8.3720  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.6540 (1.5792)  acc1: 68.2292 (69.6640)  acc5: 88.2812 (89.2480)  time: 1.0370  data: 0.9400  max mem: 21002
Test: Total time: 0:00:09 (1.0473 s / it)
* Acc@1 69.182 Acc@5 89.416 loss 1.582
Accuracy of the model on the 50000 test images: 69.2%
Max accuracy: 70.64%
Epoch: [240]  [  0/312]  eta: 0:59:51  lr: 0.000437  min_lr: 0.000437  loss: 4.3861 (4.3861)  weight_decay: 0.0500 (0.0500)  time: 11.5111  data: 8.9708  max mem: 21002
Epoch: [240]  [ 10/312]  eta: 0:07:27  lr: 0.000437  min_lr: 0.000437  loss: 3.8459 (3.6848)  weight_decay: 0.0500 (0.0500)  time: 1.4813  data: 0.9573  max mem: 21002
Epoch: [240]  [ 20/312]  eta: 0:05:09  lr: 0.000436  min_lr: 0.000436  loss: 3.5274 (3.5959)  weight_decay: 0.0500 (0.0500)  time: 0.5377  data: 0.0935  max mem: 21002
Epoch: [240]  [ 30/312]  eta: 0:03:58  lr: 0.000436  min_lr: 0.000436  loss: 3.3835 (3.4945)  weight_decay: 0.0500 (0.0500)  time: 0.4985  data: 0.0717  max mem: 21002
Epoch: [240]  [ 40/312]  eta: 0:03:35  lr: 0.000435  min_lr: 0.000435  loss: 3.6080 (3.5557)  weight_decay: 0.0500 (0.0500)  time: 0.5131  data: 0.0726  max mem: 21002
Epoch: [240]  [ 50/312]  eta: 0:03:20  lr: 0.000435  min_lr: 0.000435  loss: 3.7189 (3.5936)  weight_decay: 0.0500 (0.0500)  time: 0.6350  data: 0.0665  max mem: 21002
Epoch: [240]  [ 60/312]  eta: 0:02:52  lr: 0.000435  min_lr: 0.000435  loss: 3.7170 (3.5953)  weight_decay: 0.0500 (0.0500)  time: 0.4670  data: 0.0512  max mem: 21002
Epoch: [240]  [ 70/312]  eta: 0:02:46  lr: 0.000434  min_lr: 0.000434  loss: 3.4886 (3.5783)  weight_decay: 0.0500 (0.0500)  time: 0.4956  data: 0.0748  max mem: 21002
Epoch: [240]  [ 80/312]  eta: 0:02:34  lr: 0.000434  min_lr: 0.000434  loss: 3.4510 (3.5875)  weight_decay: 0.0500 (0.0500)  time: 0.6035  data: 0.0741  max mem: 21002
Epoch: [240]  [ 90/312]  eta: 0:02:21  lr: 0.000433  min_lr: 0.000433  loss: 3.8141 (3.6199)  weight_decay: 0.0500 (0.0500)  time: 0.4483  data: 0.0393  max mem: 21002
Epoch: [240]  [100/312]  eta: 0:02:15  lr: 0.000433  min_lr: 0.000433  loss: 3.8633 (3.6368)  weight_decay: 0.0500 (0.0500)  time: 0.5403  data: 0.0981  max mem: 21002
Epoch: [240]  [110/312]  eta: 0:02:03  lr: 0.000432  min_lr: 0.000432  loss: 3.8599 (3.6289)  weight_decay: 0.0500 (0.0500)  time: 0.4903  data: 0.0630  max mem: 21002
Epoch: [240]  [120/312]  eta: 0:01:58  lr: 0.000432  min_lr: 0.000432  loss: 3.4771 (3.6145)  weight_decay: 0.0500 (0.0500)  time: 0.4830  data: 0.0832  max mem: 21002
Epoch: [240]  [130/312]  eta: 0:01:51  lr: 0.000431  min_lr: 0.000431  loss: 3.7921 (3.6357)  weight_decay: 0.0500 (0.0500)  time: 0.6145  data: 0.1446  max mem: 21002
Epoch: [240]  [140/312]  eta: 0:01:41  lr: 0.000431  min_lr: 0.000431  loss: 3.7665 (3.6371)  weight_decay: 0.0500 (0.0500)  time: 0.4228  data: 0.0697  max mem: 21002
Epoch: [240]  [150/312]  eta: 0:01:37  lr: 0.000431  min_lr: 0.000431  loss: 3.7648 (3.6445)  weight_decay: 0.0500 (0.0500)  time: 0.5293  data: 0.0933  max mem: 21002
Epoch: [240]  [160/312]  eta: 0:01:30  lr: 0.000430  min_lr: 0.000430  loss: 3.9287 (3.6476)  weight_decay: 0.0500 (0.0500)  time: 0.6613  data: 0.0923  max mem: 21002
Epoch: [240]  [170/312]  eta: 0:01:23  lr: 0.000430  min_lr: 0.000430  loss: 3.9229 (3.6576)  weight_decay: 0.0500 (0.0500)  time: 0.4834  data: 0.0688  max mem: 21002
Epoch: [240]  [180/312]  eta: 0:01:18  lr: 0.000429  min_lr: 0.000429  loss: 3.8297 (3.6483)  weight_decay: 0.0500 (0.0500)  time: 0.5550  data: 0.1109  max mem: 21002
Epoch: [240]  [190/312]  eta: 0:01:10  lr: 0.000429  min_lr: 0.000429  loss: 3.5116 (3.6456)  weight_decay: 0.0500 (0.0500)  time: 0.5000  data: 0.0501  max mem: 21002
Epoch: [240]  [200/312]  eta: 0:01:04  lr: 0.000428  min_lr: 0.000428  loss: 3.7797 (3.6451)  weight_decay: 0.0500 (0.0500)  time: 0.4699  data: 0.0685  max mem: 21002
Epoch: [240]  [210/312]  eta: 0:00:59  lr: 0.000428  min_lr: 0.000428  loss: 3.7949 (3.6459)  weight_decay: 0.0500 (0.0500)  time: 0.6543  data: 0.1349  max mem: 21002
Epoch: [240]  [220/312]  eta: 0:00:52  lr: 0.000427  min_lr: 0.000427  loss: 3.4958 (3.6304)  weight_decay: 0.0500 (0.0500)  time: 0.5057  data: 0.0980  max mem: 21002
Epoch: [240]  [230/312]  eta: 0:00:47  lr: 0.000427  min_lr: 0.000427  loss: 3.2848 (3.6317)  weight_decay: 0.0500 (0.0500)  time: 0.5056  data: 0.0739  max mem: 21002
Epoch: [240]  [240/312]  eta: 0:00:41  lr: 0.000427  min_lr: 0.000427  loss: 3.7943 (3.6361)  weight_decay: 0.0500 (0.0500)  time: 0.5624  data: 0.0472  max mem: 21002
Epoch: [240]  [250/312]  eta: 0:00:35  lr: 0.000426  min_lr: 0.000426  loss: 3.8418 (3.6410)  weight_decay: 0.0500 (0.0500)  time: 0.4687  data: 0.0582  max mem: 21002
Epoch: [240]  [260/312]  eta: 0:00:29  lr: 0.000426  min_lr: 0.000426  loss: 3.8418 (3.6476)  weight_decay: 0.0500 (0.0500)  time: 0.5517  data: 0.0932  max mem: 21002
Epoch: [240]  [270/312]  eta: 0:00:23  lr: 0.000425  min_lr: 0.000425  loss: 3.8035 (3.6470)  weight_decay: 0.0500 (0.0500)  time: 0.4660  data: 0.0466  max mem: 21002
Epoch: [240]  [280/312]  eta: 0:00:18  lr: 0.000425  min_lr: 0.000425  loss: 3.7127 (3.6509)  weight_decay: 0.0500 (0.0500)  time: 0.5261  data: 0.0913  max mem: 21002
Epoch: [240]  [290/312]  eta: 0:00:12  lr: 0.000424  min_lr: 0.000424  loss: 3.6267 (3.6417)  weight_decay: 0.0500 (0.0500)  time: 0.5986  data: 0.0809  max mem: 21002
Epoch: [240]  [300/312]  eta: 0:00:06  lr: 0.000424  min_lr: 0.000424  loss: 3.3689 (3.6320)  weight_decay: 0.0500 (0.0500)  time: 0.4064  data: 0.0387  max mem: 21002
Epoch: [240]  [310/312]  eta: 0:00:01  lr: 0.000423  min_lr: 0.000423  loss: 3.6239 (3.6406)  weight_decay: 0.0500 (0.0500)  time: 0.3169  data: 0.0380  max mem: 21002
Epoch: [240]  [311/312]  eta: 0:00:00  lr: 0.000423  min_lr: 0.000423  loss: 3.6239 (3.6369)  weight_decay: 0.0500 (0.0500)  time: 0.3168  data: 0.0380  max mem: 21002
Epoch: [240] Total time: 0:02:51 (0.5508 s / it)
Averaged stats: lr: 0.000423  min_lr: 0.000423  loss: 3.6239 (3.6312)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.1192 (1.1192)  acc1: 79.4271 (79.4271)  acc5: 94.5312 (94.5312)  time: 8.3596  data: 8.2407  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.5143 (1.4183)  acc1: 70.9635 (71.3600)  acc5: 90.6250 (90.3840)  time: 1.0138  data: 0.9157  max mem: 21002
Test: Total time: 0:00:09 (1.0342 s / it)
* Acc@1 70.506 Acc@5 90.368 loss 1.430
Accuracy of the model on the 50000 test images: 70.5%
Max accuracy: 70.64%
Epoch: [241]  [  0/312]  eta: 1:00:17  lr: 0.000423  min_lr: 0.000423  loss: 3.4145 (3.4145)  weight_decay: 0.0500 (0.0500)  time: 11.5954  data: 9.3996  max mem: 21002
Epoch: [241]  [ 10/312]  eta: 0:08:52  lr: 0.000423  min_lr: 0.000423  loss: 3.8375 (3.6056)  weight_decay: 0.0500 (0.0500)  time: 1.7617  data: 1.0165  max mem: 21002
Epoch: [241]  [ 20/312]  eta: 0:05:16  lr: 0.000422  min_lr: 0.000422  loss: 3.9036 (3.6421)  weight_decay: 0.0500 (0.0500)  time: 0.5597  data: 0.1172  max mem: 21002
Epoch: [241]  [ 30/312]  eta: 0:03:53  lr: 0.000422  min_lr: 0.000422  loss: 3.6260 (3.5254)  weight_decay: 0.0500 (0.0500)  time: 0.3163  data: 0.0284  max mem: 21002
Epoch: [241]  [ 40/312]  eta: 0:03:09  lr: 0.000422  min_lr: 0.000422  loss: 3.5420 (3.5895)  weight_decay: 0.0500 (0.0500)  time: 0.2913  data: 0.0010  max mem: 21002
Epoch: [241]  [ 50/312]  eta: 0:02:50  lr: 0.000421  min_lr: 0.000421  loss: 3.9582 (3.6567)  weight_decay: 0.0500 (0.0500)  time: 0.3790  data: 0.0237  max mem: 21002
Epoch: [241]  [ 60/312]  eta: 0:02:36  lr: 0.000421  min_lr: 0.000421  loss: 3.9604 (3.6194)  weight_decay: 0.0500 (0.0500)  time: 0.4634  data: 0.1013  max mem: 21002
Epoch: [241]  [ 70/312]  eta: 0:02:32  lr: 0.000420  min_lr: 0.000420  loss: 3.9780 (3.6533)  weight_decay: 0.0500 (0.0500)  time: 0.5691  data: 0.1443  max mem: 21002
Epoch: [241]  [ 80/312]  eta: 0:02:20  lr: 0.000420  min_lr: 0.000420  loss: 3.8754 (3.6273)  weight_decay: 0.0500 (0.0500)  time: 0.5549  data: 0.1026  max mem: 21002
Epoch: [241]  [ 90/312]  eta: 0:02:14  lr: 0.000419  min_lr: 0.000419  loss: 3.3421 (3.6086)  weight_decay: 0.0500 (0.0500)  time: 0.5142  data: 0.1179  max mem: 21002
Epoch: [241]  [100/312]  eta: 0:02:08  lr: 0.000419  min_lr: 0.000419  loss: 3.8425 (3.6420)  weight_decay: 0.0500 (0.0500)  time: 0.6107  data: 0.1076  max mem: 21002
Epoch: [241]  [110/312]  eta: 0:01:58  lr: 0.000418  min_lr: 0.000418  loss: 3.9000 (3.6444)  weight_decay: 0.0500 (0.0500)  time: 0.5151  data: 0.0644  max mem: 21002
Epoch: [241]  [120/312]  eta: 0:01:53  lr: 0.000418  min_lr: 0.000418  loss: 3.7862 (3.6547)  weight_decay: 0.0500 (0.0500)  time: 0.5102  data: 0.0880  max mem: 21002
Epoch: [241]  [130/312]  eta: 0:01:44  lr: 0.000418  min_lr: 0.000418  loss: 3.8025 (3.6587)  weight_decay: 0.0500 (0.0500)  time: 0.4999  data: 0.0673  max mem: 21002
Epoch: [241]  [140/312]  eta: 0:01:39  lr: 0.000417  min_lr: 0.000417  loss: 3.8234 (3.6638)  weight_decay: 0.0500 (0.0500)  time: 0.5201  data: 0.0909  max mem: 21002
Epoch: [241]  [150/312]  eta: 0:01:34  lr: 0.000417  min_lr: 0.000417  loss: 3.8841 (3.6633)  weight_decay: 0.0500 (0.0500)  time: 0.6432  data: 0.1197  max mem: 21002
Epoch: [241]  [160/312]  eta: 0:01:26  lr: 0.000416  min_lr: 0.000416  loss: 3.6726 (3.6534)  weight_decay: 0.0500 (0.0500)  time: 0.4852  data: 0.0716  max mem: 21002
Epoch: [241]  [170/312]  eta: 0:01:22  lr: 0.000416  min_lr: 0.000416  loss: 3.8075 (3.6691)  weight_decay: 0.0500 (0.0500)  time: 0.5475  data: 0.0895  max mem: 21002
Epoch: [241]  [180/312]  eta: 0:01:16  lr: 0.000415  min_lr: 0.000415  loss: 3.7998 (3.6580)  weight_decay: 0.0500 (0.0500)  time: 0.6677  data: 0.1035  max mem: 21002
Epoch: [241]  [190/312]  eta: 0:01:09  lr: 0.000415  min_lr: 0.000415  loss: 3.7558 (3.6653)  weight_decay: 0.0500 (0.0500)  time: 0.4864  data: 0.0630  max mem: 21002
Epoch: [241]  [200/312]  eta: 0:01:04  lr: 0.000415  min_lr: 0.000415  loss: 3.8921 (3.6684)  weight_decay: 0.0500 (0.0500)  time: 0.5468  data: 0.0683  max mem: 21002
Epoch: [241]  [210/312]  eta: 0:00:57  lr: 0.000414  min_lr: 0.000414  loss: 3.8737 (3.6625)  weight_decay: 0.0500 (0.0500)  time: 0.5051  data: 0.0452  max mem: 21002
Epoch: [241]  [220/312]  eta: 0:00:52  lr: 0.000414  min_lr: 0.000414  loss: 2.9625 (3.6411)  weight_decay: 0.0500 (0.0500)  time: 0.5263  data: 0.0733  max mem: 21002
Epoch: [241]  [230/312]  eta: 0:00:46  lr: 0.000413  min_lr: 0.000413  loss: 3.3413 (3.6476)  weight_decay: 0.0500 (0.0500)  time: 0.6597  data: 0.0731  max mem: 21002
Epoch: [241]  [240/312]  eta: 0:00:40  lr: 0.000413  min_lr: 0.000413  loss: 3.8828 (3.6466)  weight_decay: 0.0500 (0.0500)  time: 0.4784  data: 0.0304  max mem: 21002
Epoch: [241]  [250/312]  eta: 0:00:35  lr: 0.000412  min_lr: 0.000412  loss: 3.6271 (3.6441)  weight_decay: 0.0500 (0.0500)  time: 0.5204  data: 0.0671  max mem: 21002
Epoch: [241]  [260/312]  eta: 0:00:29  lr: 0.000412  min_lr: 0.000412  loss: 3.6216 (3.6420)  weight_decay: 0.0500 (0.0500)  time: 0.6076  data: 0.0374  max mem: 21002
Epoch: [241]  [270/312]  eta: 0:00:23  lr: 0.000411  min_lr: 0.000411  loss: 3.7167 (3.6457)  weight_decay: 0.0500 (0.0500)  time: 0.5057  data: 0.0293  max mem: 21002
Epoch: [241]  [280/312]  eta: 0:00:18  lr: 0.000411  min_lr: 0.000411  loss: 3.7716 (3.6488)  weight_decay: 0.0500 (0.0500)  time: 0.6056  data: 0.0791  max mem: 21002
Epoch: [241]  [290/312]  eta: 0:00:12  lr: 0.000411  min_lr: 0.000411  loss: 3.8378 (3.6525)  weight_decay: 0.0500 (0.0500)  time: 0.5207  data: 0.0506  max mem: 21002
Epoch: [241]  [300/312]  eta: 0:00:06  lr: 0.000410  min_lr: 0.000410  loss: 3.9390 (3.6537)  weight_decay: 0.0500 (0.0500)  time: 0.3828  data: 0.0148  max mem: 21002
Epoch: [241]  [310/312]  eta: 0:00:01  lr: 0.000410  min_lr: 0.000410  loss: 3.6685 (3.6481)  weight_decay: 0.0500 (0.0500)  time: 0.3774  data: 0.0145  max mem: 21002
Epoch: [241]  [311/312]  eta: 0:00:00  lr: 0.000410  min_lr: 0.000410  loss: 3.8088 (3.6488)  weight_decay: 0.0500 (0.0500)  time: 0.3731  data: 0.0097  max mem: 21002
Epoch: [241] Total time: 0:02:52 (0.5514 s / it)
Averaged stats: lr: 0.000410  min_lr: 0.000410  loss: 3.8088 (3.6475)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:11  loss: 1.2249 (1.2249)  acc1: 77.3438 (77.3438)  acc5: 93.3594 (93.3594)  time: 7.9246  data: 7.8086  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.5463 (1.4651)  acc1: 68.3594 (69.7760)  acc5: 88.8021 (89.3280)  time: 1.0278  data: 0.9297  max mem: 21002
Test: Total time: 0:00:09 (1.0412 s / it)
* Acc@1 69.482 Acc@5 89.796 loss 1.468
Accuracy of the model on the 50000 test images: 69.5%
Max accuracy: 70.64%
Epoch: [242]  [  0/312]  eta: 1:06:29  lr: 0.000410  min_lr: 0.000410  loss: 2.7114 (2.7114)  weight_decay: 0.0500 (0.0500)  time: 12.7853  data: 10.2701  max mem: 21002
Epoch: [242]  [ 10/312]  eta: 0:08:20  lr: 0.000409  min_lr: 0.000409  loss: 3.8019 (3.5576)  weight_decay: 0.0500 (0.0500)  time: 1.6562  data: 1.0814  max mem: 21002
Epoch: [242]  [ 20/312]  eta: 0:05:15  lr: 0.000409  min_lr: 0.000409  loss: 3.8019 (3.5798)  weight_decay: 0.0500 (0.0500)  time: 0.4961  data: 0.1116  max mem: 21002
Epoch: [242]  [ 30/312]  eta: 0:03:53  lr: 0.000408  min_lr: 0.000408  loss: 3.7774 (3.6205)  weight_decay: 0.0500 (0.0500)  time: 0.3718  data: 0.0307  max mem: 21002
Epoch: [242]  [ 40/312]  eta: 0:03:19  lr: 0.000408  min_lr: 0.000408  loss: 3.6126 (3.5618)  weight_decay: 0.0500 (0.0500)  time: 0.3690  data: 0.0473  max mem: 21002
Epoch: [242]  [ 50/312]  eta: 0:03:10  lr: 0.000407  min_lr: 0.000407  loss: 3.4964 (3.5702)  weight_decay: 0.0500 (0.0500)  time: 0.5712  data: 0.2008  max mem: 21002
Epoch: [242]  [ 60/312]  eta: 0:02:45  lr: 0.000407  min_lr: 0.000407  loss: 3.8397 (3.6039)  weight_decay: 0.0500 (0.0500)  time: 0.4932  data: 0.1542  max mem: 21002
Epoch: [242]  [ 70/312]  eta: 0:02:39  lr: 0.000407  min_lr: 0.000407  loss: 3.8057 (3.5923)  weight_decay: 0.0500 (0.0500)  time: 0.4923  data: 0.1577  max mem: 21002
Epoch: [242]  [ 80/312]  eta: 0:02:30  lr: 0.000406  min_lr: 0.000406  loss: 3.5061 (3.5879)  weight_decay: 0.0500 (0.0500)  time: 0.6318  data: 0.2147  max mem: 21002
Epoch: [242]  [ 90/312]  eta: 0:02:18  lr: 0.000406  min_lr: 0.000406  loss: 3.8016 (3.6077)  weight_decay: 0.0500 (0.0500)  time: 0.5014  data: 0.1320  max mem: 21002
Epoch: [242]  [100/312]  eta: 0:02:13  lr: 0.000405  min_lr: 0.000405  loss: 3.6425 (3.5906)  weight_decay: 0.0500 (0.0500)  time: 0.5478  data: 0.2091  max mem: 21002
Epoch: [242]  [110/312]  eta: 0:02:00  lr: 0.000405  min_lr: 0.000405  loss: 3.6863 (3.6036)  weight_decay: 0.0500 (0.0500)  time: 0.4742  data: 0.1348  max mem: 21002
Epoch: [242]  [120/312]  eta: 0:01:56  lr: 0.000404  min_lr: 0.000404  loss: 3.8672 (3.6097)  weight_decay: 0.0500 (0.0500)  time: 0.4811  data: 0.1198  max mem: 21002
Epoch: [242]  [130/312]  eta: 0:01:50  lr: 0.000404  min_lr: 0.000404  loss: 3.9335 (3.6290)  weight_decay: 0.0500 (0.0500)  time: 0.6714  data: 0.2283  max mem: 21002
Epoch: [242]  [140/312]  eta: 0:01:40  lr: 0.000404  min_lr: 0.000404  loss: 3.9901 (3.6382)  weight_decay: 0.0500 (0.0500)  time: 0.4789  data: 0.1091  max mem: 21002
Epoch: [242]  [150/312]  eta: 0:01:36  lr: 0.000403  min_lr: 0.000403  loss: 3.5881 (3.6247)  weight_decay: 0.0500 (0.0500)  time: 0.4831  data: 0.1142  max mem: 21002
Epoch: [242]  [160/312]  eta: 0:01:29  lr: 0.000403  min_lr: 0.000403  loss: 3.6586 (3.6371)  weight_decay: 0.0500 (0.0500)  time: 0.6231  data: 0.2216  max mem: 21002
Epoch: [242]  [170/312]  eta: 0:01:22  lr: 0.000402  min_lr: 0.000402  loss: 3.6393 (3.6219)  weight_decay: 0.0500 (0.0500)  time: 0.4985  data: 0.1537  max mem: 21002
Epoch: [242]  [180/312]  eta: 0:01:17  lr: 0.000402  min_lr: 0.000402  loss: 3.4437 (3.6221)  weight_decay: 0.0500 (0.0500)  time: 0.5648  data: 0.1758  max mem: 21002
Epoch: [242]  [190/312]  eta: 0:01:09  lr: 0.000401  min_lr: 0.000401  loss: 3.9045 (3.6320)  weight_decay: 0.0500 (0.0500)  time: 0.4965  data: 0.1302  max mem: 21002
Epoch: [242]  [200/312]  eta: 0:01:04  lr: 0.000401  min_lr: 0.000401  loss: 3.7474 (3.6171)  weight_decay: 0.0500 (0.0500)  time: 0.4862  data: 0.1153  max mem: 21002
Epoch: [242]  [210/312]  eta: 0:00:59  lr: 0.000401  min_lr: 0.000401  loss: 3.4376 (3.6095)  weight_decay: 0.0500 (0.0500)  time: 0.7039  data: 0.2115  max mem: 21002
Epoch: [242]  [220/312]  eta: 0:00:52  lr: 0.000400  min_lr: 0.000400  loss: 3.8372 (3.6167)  weight_decay: 0.0500 (0.0500)  time: 0.5397  data: 0.1274  max mem: 21002
Epoch: [242]  [230/312]  eta: 0:00:47  lr: 0.000400  min_lr: 0.000400  loss: 3.8797 (3.6215)  weight_decay: 0.0500 (0.0500)  time: 0.5021  data: 0.1106  max mem: 21002
Epoch: [242]  [240/312]  eta: 0:00:41  lr: 0.000399  min_lr: 0.000399  loss: 3.8797 (3.6278)  weight_decay: 0.0500 (0.0500)  time: 0.5872  data: 0.0807  max mem: 21002
Epoch: [242]  [250/312]  eta: 0:00:35  lr: 0.000399  min_lr: 0.000399  loss: 3.6028 (3.6227)  weight_decay: 0.0500 (0.0500)  time: 0.4902  data: 0.0873  max mem: 21002
Epoch: [242]  [260/312]  eta: 0:00:30  lr: 0.000398  min_lr: 0.000398  loss: 3.6913 (3.6256)  weight_decay: 0.0500 (0.0500)  time: 0.6089  data: 0.1845  max mem: 21002
Epoch: [242]  [270/312]  eta: 0:00:23  lr: 0.000398  min_lr: 0.000398  loss: 3.7265 (3.6169)  weight_decay: 0.0500 (0.0500)  time: 0.5230  data: 0.0987  max mem: 21002
Epoch: [242]  [280/312]  eta: 0:00:18  lr: 0.000398  min_lr: 0.000398  loss: 3.7906 (3.6188)  weight_decay: 0.0500 (0.0500)  time: 0.4552  data: 0.0788  max mem: 21002
Epoch: [242]  [290/312]  eta: 0:00:12  lr: 0.000397  min_lr: 0.000397  loss: 3.8945 (3.6260)  weight_decay: 0.0500 (0.0500)  time: 0.5653  data: 0.1186  max mem: 21002
Epoch: [242]  [300/312]  eta: 0:00:06  lr: 0.000397  min_lr: 0.000397  loss: 3.9983 (3.6295)  weight_decay: 0.0500 (0.0500)  time: 0.4061  data: 0.0541  max mem: 21002
Epoch: [242]  [310/312]  eta: 0:00:01  lr: 0.000396  min_lr: 0.000396  loss: 4.0274 (3.6406)  weight_decay: 0.0500 (0.0500)  time: 0.2908  data: 0.0140  max mem: 21002
Epoch: [242]  [311/312]  eta: 0:00:00  lr: 0.000396  min_lr: 0.000396  loss: 4.0189 (3.6415)  weight_decay: 0.0500 (0.0500)  time: 0.2906  data: 0.0140  max mem: 21002
Epoch: [242] Total time: 0:02:52 (0.5524 s / it)
Averaged stats: lr: 0.000396  min_lr: 0.000396  loss: 4.0189 (3.6635)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.2456 (1.2456)  acc1: 79.2969 (79.2969)  acc5: 94.1406 (94.1406)  time: 8.7366  data: 8.6183  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.5791 (1.5325)  acc1: 71.0938 (71.4560)  acc5: 90.8854 (90.5120)  time: 1.0631  data: 0.9577  max mem: 21002
Test: Total time: 0:00:09 (1.0737 s / it)
* Acc@1 70.722 Acc@5 90.362 loss 1.543
Accuracy of the model on the 50000 test images: 70.7%
Max accuracy: 70.72%
Epoch: [243]  [  0/312]  eta: 1:01:20  lr: 0.000396  min_lr: 0.000396  loss: 3.8842 (3.8842)  weight_decay: 0.0500 (0.0500)  time: 11.7968  data: 10.4995  max mem: 21002
Epoch: [243]  [ 10/312]  eta: 0:08:04  lr: 0.000396  min_lr: 0.000396  loss: 3.8842 (3.8028)  weight_decay: 0.0500 (0.0500)  time: 1.6040  data: 1.0660  max mem: 21002
Epoch: [243]  [ 20/312]  eta: 0:05:04  lr: 0.000395  min_lr: 0.000395  loss: 3.8512 (3.7126)  weight_decay: 0.0500 (0.0500)  time: 0.5034  data: 0.1141  max mem: 21002
Epoch: [243]  [ 30/312]  eta: 0:03:45  lr: 0.000395  min_lr: 0.000395  loss: 3.7218 (3.6984)  weight_decay: 0.0500 (0.0500)  time: 0.3551  data: 0.0533  max mem: 21002
Epoch: [243]  [ 40/312]  eta: 0:03:19  lr: 0.000394  min_lr: 0.000394  loss: 3.6820 (3.6725)  weight_decay: 0.0500 (0.0500)  time: 0.4095  data: 0.0810  max mem: 21002
Epoch: [243]  [ 50/312]  eta: 0:03:09  lr: 0.000394  min_lr: 0.000394  loss: 3.6652 (3.6624)  weight_decay: 0.0500 (0.0500)  time: 0.6089  data: 0.2104  max mem: 21002
Epoch: [243]  [ 60/312]  eta: 0:02:44  lr: 0.000394  min_lr: 0.000394  loss: 3.6652 (3.6529)  weight_decay: 0.0500 (0.0500)  time: 0.4878  data: 0.1303  max mem: 21002
Epoch: [243]  [ 70/312]  eta: 0:02:39  lr: 0.000393  min_lr: 0.000393  loss: 3.7754 (3.6444)  weight_decay: 0.0500 (0.0500)  time: 0.5008  data: 0.1314  max mem: 21002
Epoch: [243]  [ 80/312]  eta: 0:02:29  lr: 0.000393  min_lr: 0.000393  loss: 3.8310 (3.6625)  weight_decay: 0.0500 (0.0500)  time: 0.6174  data: 0.2486  max mem: 21002
Epoch: [243]  [ 90/312]  eta: 0:02:18  lr: 0.000392  min_lr: 0.000392  loss: 3.7808 (3.6567)  weight_decay: 0.0500 (0.0500)  time: 0.4890  data: 0.1181  max mem: 21002
Epoch: [243]  [100/312]  eta: 0:02:13  lr: 0.000392  min_lr: 0.000392  loss: 3.7885 (3.6726)  weight_decay: 0.0500 (0.0500)  time: 0.5836  data: 0.1201  max mem: 21002
Epoch: [243]  [110/312]  eta: 0:02:01  lr: 0.000391  min_lr: 0.000391  loss: 3.7885 (3.6743)  weight_decay: 0.0500 (0.0500)  time: 0.4983  data: 0.1199  max mem: 21002
Epoch: [243]  [120/312]  eta: 0:01:56  lr: 0.000391  min_lr: 0.000391  loss: 3.7267 (3.6891)  weight_decay: 0.0500 (0.0500)  time: 0.4838  data: 0.0993  max mem: 21002
Epoch: [243]  [130/312]  eta: 0:01:52  lr: 0.000391  min_lr: 0.000391  loss: 3.8667 (3.6962)  weight_decay: 0.0500 (0.0500)  time: 0.7008  data: 0.2187  max mem: 21002
Epoch: [243]  [140/312]  eta: 0:01:41  lr: 0.000390  min_lr: 0.000390  loss: 3.9154 (3.6963)  weight_decay: 0.0500 (0.0500)  time: 0.5030  data: 0.1202  max mem: 21002
Epoch: [243]  [150/312]  eta: 0:01:37  lr: 0.000390  min_lr: 0.000390  loss: 3.8673 (3.7006)  weight_decay: 0.0500 (0.0500)  time: 0.4979  data: 0.0944  max mem: 21002
Epoch: [243]  [160/312]  eta: 0:01:29  lr: 0.000389  min_lr: 0.000389  loss: 4.0387 (3.7135)  weight_decay: 0.0500 (0.0500)  time: 0.5775  data: 0.1729  max mem: 21002
Epoch: [243]  [170/312]  eta: 0:01:23  lr: 0.000389  min_lr: 0.000389  loss: 3.9517 (3.7248)  weight_decay: 0.0500 (0.0500)  time: 0.4964  data: 0.0792  max mem: 21002
Epoch: [243]  [180/312]  eta: 0:01:18  lr: 0.000388  min_lr: 0.000388  loss: 3.9517 (3.7329)  weight_decay: 0.0500 (0.0500)  time: 0.6089  data: 0.0732  max mem: 21002
Epoch: [243]  [190/312]  eta: 0:01:10  lr: 0.000388  min_lr: 0.000388  loss: 3.9482 (3.7360)  weight_decay: 0.0500 (0.0500)  time: 0.4846  data: 0.0732  max mem: 21002
Epoch: [243]  [200/312]  eta: 0:01:05  lr: 0.000388  min_lr: 0.000388  loss: 3.6022 (3.7218)  weight_decay: 0.0500 (0.0500)  time: 0.4983  data: 0.0751  max mem: 21002
Epoch: [243]  [210/312]  eta: 0:01:00  lr: 0.000387  min_lr: 0.000387  loss: 3.5013 (3.7254)  weight_decay: 0.0500 (0.0500)  time: 0.6990  data: 0.1082  max mem: 21002
Epoch: [243]  [220/312]  eta: 0:00:52  lr: 0.000387  min_lr: 0.000387  loss: 3.6344 (3.7222)  weight_decay: 0.0500 (0.0500)  time: 0.4997  data: 0.0338  max mem: 21002
Epoch: [243]  [230/312]  eta: 0:00:47  lr: 0.000386  min_lr: 0.000386  loss: 3.8735 (3.7235)  weight_decay: 0.0500 (0.0500)  time: 0.5007  data: 0.0477  max mem: 21002
Epoch: [243]  [240/312]  eta: 0:00:41  lr: 0.000386  min_lr: 0.000386  loss: 3.8735 (3.7237)  weight_decay: 0.0500 (0.0500)  time: 0.5592  data: 0.1076  max mem: 21002
Epoch: [243]  [250/312]  eta: 0:00:35  lr: 0.000385  min_lr: 0.000385  loss: 3.6326 (3.7100)  weight_decay: 0.0500 (0.0500)  time: 0.5135  data: 0.0606  max mem: 21002
Epoch: [243]  [260/312]  eta: 0:00:30  lr: 0.000385  min_lr: 0.000385  loss: 3.3128 (3.7010)  weight_decay: 0.0500 (0.0500)  time: 0.6606  data: 0.0513  max mem: 21002
Epoch: [243]  [270/312]  eta: 0:00:23  lr: 0.000385  min_lr: 0.000385  loss: 3.5447 (3.6985)  weight_decay: 0.0500 (0.0500)  time: 0.5023  data: 0.0518  max mem: 21002
Epoch: [243]  [280/312]  eta: 0:00:18  lr: 0.000384  min_lr: 0.000384  loss: 3.6267 (3.7001)  weight_decay: 0.0500 (0.0500)  time: 0.4844  data: 0.0215  max mem: 21002
Epoch: [243]  [290/312]  eta: 0:00:12  lr: 0.000384  min_lr: 0.000384  loss: 3.9153 (3.7048)  weight_decay: 0.0500 (0.0500)  time: 0.5832  data: 0.0594  max mem: 21002
Epoch: [243]  [300/312]  eta: 0:00:06  lr: 0.000383  min_lr: 0.000383  loss: 3.9428 (3.7027)  weight_decay: 0.0500 (0.0500)  time: 0.3842  data: 0.0389  max mem: 21002
Epoch: [243]  [310/312]  eta: 0:00:01  lr: 0.000383  min_lr: 0.000383  loss: 3.8223 (3.7029)  weight_decay: 0.0500 (0.0500)  time: 0.2768  data: 0.0001  max mem: 21002
Epoch: [243]  [311/312]  eta: 0:00:00  lr: 0.000383  min_lr: 0.000383  loss: 3.8223 (3.7030)  weight_decay: 0.0500 (0.0500)  time: 0.2768  data: 0.0001  max mem: 21002
Epoch: [243] Total time: 0:02:53 (0.5553 s / it)
Averaged stats: lr: 0.000383  min_lr: 0.000383  loss: 3.8223 (3.6605)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.1596 (1.1596)  acc1: 79.1667 (79.1667)  acc5: 94.9219 (94.9219)  time: 8.8149  data: 8.6964  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.5027 (1.4544)  acc1: 70.5729 (70.9760)  acc5: 90.4948 (90.3520)  time: 1.0680  data: 0.9663  max mem: 21002
Test: Total time: 0:00:09 (1.0821 s / it)
* Acc@1 70.678 Acc@5 90.356 loss 1.461
Accuracy of the model on the 50000 test images: 70.7%
Max accuracy: 70.72%
Epoch: [244]  [  0/312]  eta: 1:01:57  lr: 0.000383  min_lr: 0.000383  loss: 3.6579 (3.6579)  weight_decay: 0.0500 (0.0500)  time: 11.9137  data: 11.5318  max mem: 21002
Epoch: [244]  [ 10/312]  eta: 0:08:38  lr: 0.000382  min_lr: 0.000382  loss: 4.0683 (3.8333)  weight_decay: 0.0500 (0.0500)  time: 1.7157  data: 1.0489  max mem: 21002
Epoch: [244]  [ 20/312]  eta: 0:05:08  lr: 0.000382  min_lr: 0.000382  loss: 3.6716 (3.7002)  weight_decay: 0.0500 (0.0500)  time: 0.5132  data: 0.0126  max mem: 21002
Epoch: [244]  [ 30/312]  eta: 0:03:48  lr: 0.000382  min_lr: 0.000382  loss: 3.6716 (3.6699)  weight_decay: 0.0500 (0.0500)  time: 0.3119  data: 0.0133  max mem: 21002
Epoch: [244]  [ 40/312]  eta: 0:03:05  lr: 0.000381  min_lr: 0.000381  loss: 3.6941 (3.6412)  weight_decay: 0.0500 (0.0500)  time: 0.2927  data: 0.0014  max mem: 21002
Epoch: [244]  [ 50/312]  eta: 0:02:55  lr: 0.000381  min_lr: 0.000381  loss: 3.2977 (3.6131)  weight_decay: 0.0500 (0.0500)  time: 0.4486  data: 0.1586  max mem: 21002
Epoch: [244]  [ 60/312]  eta: 0:02:33  lr: 0.000380  min_lr: 0.000380  loss: 3.2867 (3.5834)  weight_decay: 0.0500 (0.0500)  time: 0.4512  data: 0.1586  max mem: 21002
Epoch: [244]  [ 70/312]  eta: 0:02:27  lr: 0.000380  min_lr: 0.000380  loss: 3.7589 (3.6447)  weight_decay: 0.0500 (0.0500)  time: 0.4614  data: 0.1277  max mem: 21002
Epoch: [244]  [ 80/312]  eta: 0:02:23  lr: 0.000379  min_lr: 0.000379  loss: 3.9779 (3.6424)  weight_decay: 0.0500 (0.0500)  time: 0.6438  data: 0.2206  max mem: 21002
Epoch: [244]  [ 90/312]  eta: 0:02:11  lr: 0.000379  min_lr: 0.000379  loss: 3.5315 (3.6579)  weight_decay: 0.0500 (0.0500)  time: 0.5250  data: 0.1441  max mem: 21002
Epoch: [244]  [100/312]  eta: 0:02:06  lr: 0.000379  min_lr: 0.000379  loss: 3.9188 (3.6608)  weight_decay: 0.0500 (0.0500)  time: 0.5119  data: 0.1660  max mem: 21002
Epoch: [244]  [110/312]  eta: 0:01:55  lr: 0.000378  min_lr: 0.000378  loss: 3.9153 (3.6800)  weight_decay: 0.0500 (0.0500)  time: 0.4929  data: 0.1411  max mem: 21002
Epoch: [244]  [120/312]  eta: 0:01:50  lr: 0.000378  min_lr: 0.000378  loss: 3.8296 (3.6802)  weight_decay: 0.0500 (0.0500)  time: 0.4804  data: 0.1250  max mem: 21002
Epoch: [244]  [130/312]  eta: 0:01:46  lr: 0.000377  min_lr: 0.000377  loss: 3.4361 (3.6560)  weight_decay: 0.0500 (0.0500)  time: 0.6581  data: 0.2298  max mem: 21002
Epoch: [244]  [140/312]  eta: 0:01:37  lr: 0.000377  min_lr: 0.000377  loss: 3.2997 (3.6389)  weight_decay: 0.0500 (0.0500)  time: 0.5179  data: 0.1316  max mem: 21002
Epoch: [244]  [150/312]  eta: 0:01:32  lr: 0.000377  min_lr: 0.000377  loss: 3.7578 (3.6570)  weight_decay: 0.0500 (0.0500)  time: 0.4848  data: 0.1073  max mem: 21002
Epoch: [244]  [160/312]  eta: 0:01:26  lr: 0.000376  min_lr: 0.000376  loss: 3.7578 (3.6334)  weight_decay: 0.0500 (0.0500)  time: 0.5966  data: 0.2316  max mem: 21002
Epoch: [244]  [170/312]  eta: 0:01:20  lr: 0.000376  min_lr: 0.000376  loss: 3.4271 (3.6390)  weight_decay: 0.0500 (0.0500)  time: 0.4976  data: 0.1258  max mem: 21002
Epoch: [244]  [180/312]  eta: 0:01:15  lr: 0.000375  min_lr: 0.000375  loss: 3.6087 (3.6361)  weight_decay: 0.0500 (0.0500)  time: 0.5492  data: 0.1112  max mem: 21002
Epoch: [244]  [190/312]  eta: 0:01:08  lr: 0.000375  min_lr: 0.000375  loss: 3.6087 (3.6347)  weight_decay: 0.0500 (0.0500)  time: 0.5085  data: 0.1203  max mem: 21002
Epoch: [244]  [200/312]  eta: 0:01:03  lr: 0.000374  min_lr: 0.000374  loss: 3.9403 (3.6425)  weight_decay: 0.0500 (0.0500)  time: 0.5140  data: 0.1273  max mem: 21002
Epoch: [244]  [210/312]  eta: 0:00:57  lr: 0.000374  min_lr: 0.000374  loss: 3.9042 (3.6403)  weight_decay: 0.0500 (0.0500)  time: 0.6447  data: 0.2385  max mem: 21002
Epoch: [244]  [220/312]  eta: 0:00:51  lr: 0.000374  min_lr: 0.000374  loss: 3.6212 (3.6203)  weight_decay: 0.0500 (0.0500)  time: 0.4882  data: 0.1211  max mem: 21002
Epoch: [244]  [230/312]  eta: 0:00:46  lr: 0.000373  min_lr: 0.000373  loss: 3.3295 (3.6205)  weight_decay: 0.0500 (0.0500)  time: 0.5200  data: 0.1237  max mem: 21002
Epoch: [244]  [240/312]  eta: 0:00:40  lr: 0.000373  min_lr: 0.000373  loss: 3.7116 (3.6287)  weight_decay: 0.0500 (0.0500)  time: 0.6306  data: 0.2486  max mem: 21002
Epoch: [244]  [250/312]  eta: 0:00:34  lr: 0.000372  min_lr: 0.000372  loss: 3.8847 (3.6354)  weight_decay: 0.0500 (0.0500)  time: 0.5021  data: 0.1257  max mem: 21002
Epoch: [244]  [260/312]  eta: 0:00:29  lr: 0.000372  min_lr: 0.000372  loss: 3.7481 (3.6362)  weight_decay: 0.0500 (0.0500)  time: 0.4948  data: 0.1245  max mem: 21002
Epoch: [244]  [270/312]  eta: 0:00:23  lr: 0.000372  min_lr: 0.000372  loss: 3.5594 (3.6283)  weight_decay: 0.0500 (0.0500)  time: 0.4875  data: 0.1245  max mem: 21002
Epoch: [244]  [280/312]  eta: 0:00:17  lr: 0.000371  min_lr: 0.000371  loss: 3.7089 (3.6329)  weight_decay: 0.0500 (0.0500)  time: 0.5122  data: 0.1382  max mem: 21002
Epoch: [244]  [290/312]  eta: 0:00:12  lr: 0.000371  min_lr: 0.000371  loss: 3.8204 (3.6397)  weight_decay: 0.0500 (0.0500)  time: 0.6511  data: 0.2906  max mem: 21002
Epoch: [244]  [300/312]  eta: 0:00:06  lr: 0.000370  min_lr: 0.000370  loss: 3.8204 (3.6328)  weight_decay: 0.0500 (0.0500)  time: 0.4851  data: 0.1529  max mem: 21002
Epoch: [244]  [310/312]  eta: 0:00:01  lr: 0.000370  min_lr: 0.000370  loss: 3.3553 (3.6255)  weight_decay: 0.0500 (0.0500)  time: 0.3143  data: 0.0310  max mem: 21002
Epoch: [244]  [311/312]  eta: 0:00:00  lr: 0.000370  min_lr: 0.000370  loss: 3.3619 (3.6257)  weight_decay: 0.0500 (0.0500)  time: 0.3140  data: 0.0310  max mem: 21002
Epoch: [244] Total time: 0:02:50 (0.5465 s / it)
Averaged stats: lr: 0.000370  min_lr: 0.000370  loss: 3.3619 (3.6510)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.1336 (1.1336)  acc1: 79.0365 (79.0365)  acc5: 94.5312 (94.5312)  time: 8.8020  data: 8.6833  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4598 (1.3654)  acc1: 70.5729 (71.7120)  acc5: 89.9740 (90.5440)  time: 1.0631  data: 0.9649  max mem: 21002
Test: Total time: 0:00:09 (1.0888 s / it)
* Acc@1 71.344 Acc@5 90.510 loss 1.373
Accuracy of the model on the 50000 test images: 71.3%
Max accuracy: 71.34%
Epoch: [245]  [  0/312]  eta: 1:00:33  lr: 0.000370  min_lr: 0.000370  loss: 3.5391 (3.5391)  weight_decay: 0.0500 (0.0500)  time: 11.6447  data: 9.2924  max mem: 21002
Epoch: [245]  [ 10/312]  eta: 0:07:37  lr: 0.000369  min_lr: 0.000369  loss: 3.6434 (3.6717)  weight_decay: 0.0500 (0.0500)  time: 1.5146  data: 1.0077  max mem: 21002
Epoch: [245]  [ 20/312]  eta: 0:04:51  lr: 0.000369  min_lr: 0.000369  loss: 3.5752 (3.5458)  weight_decay: 0.0500 (0.0500)  time: 0.4666  data: 0.1363  max mem: 21002
Epoch: [245]  [ 30/312]  eta: 0:03:37  lr: 0.000369  min_lr: 0.000369  loss: 3.5354 (3.5537)  weight_decay: 0.0500 (0.0500)  time: 0.3626  data: 0.0521  max mem: 21002
Epoch: [245]  [ 40/312]  eta: 0:03:13  lr: 0.000368  min_lr: 0.000368  loss: 3.5354 (3.5372)  weight_decay: 0.0500 (0.0500)  time: 0.4102  data: 0.0928  max mem: 21002
Epoch: [245]  [ 50/312]  eta: 0:03:10  lr: 0.000368  min_lr: 0.000368  loss: 3.7071 (3.5838)  weight_decay: 0.0500 (0.0500)  time: 0.6588  data: 0.2394  max mem: 21002
Epoch: [245]  [ 60/312]  eta: 0:02:44  lr: 0.000367  min_lr: 0.000367  loss: 3.8285 (3.6057)  weight_decay: 0.0500 (0.0500)  time: 0.5366  data: 0.1524  max mem: 21002
Epoch: [245]  [ 70/312]  eta: 0:02:39  lr: 0.000367  min_lr: 0.000367  loss: 3.7084 (3.6132)  weight_decay: 0.0500 (0.0500)  time: 0.4870  data: 0.0684  max mem: 21002
Epoch: [245]  [ 80/312]  eta: 0:02:29  lr: 0.000366  min_lr: 0.000366  loss: 3.8080 (3.6391)  weight_decay: 0.0500 (0.0500)  time: 0.6086  data: 0.1063  max mem: 21002
Epoch: [245]  [ 90/312]  eta: 0:02:19  lr: 0.000366  min_lr: 0.000366  loss: 3.8152 (3.6485)  weight_decay: 0.0500 (0.0500)  time: 0.5127  data: 0.1347  max mem: 21002
Epoch: [245]  [100/312]  eta: 0:02:14  lr: 0.000366  min_lr: 0.000366  loss: 3.6288 (3.6376)  weight_decay: 0.0500 (0.0500)  time: 0.5961  data: 0.2017  max mem: 21002
Epoch: [245]  [110/312]  eta: 0:02:01  lr: 0.000365  min_lr: 0.000365  loss: 3.6288 (3.6310)  weight_decay: 0.0500 (0.0500)  time: 0.4939  data: 0.1056  max mem: 21002
Epoch: [245]  [120/312]  eta: 0:01:56  lr: 0.000365  min_lr: 0.000365  loss: 3.7600 (3.6289)  weight_decay: 0.0500 (0.0500)  time: 0.4673  data: 0.0829  max mem: 21002
Epoch: [245]  [130/312]  eta: 0:01:51  lr: 0.000364  min_lr: 0.000364  loss: 3.5472 (3.6002)  weight_decay: 0.0500 (0.0500)  time: 0.6660  data: 0.1940  max mem: 21002
Epoch: [245]  [140/312]  eta: 0:01:41  lr: 0.000364  min_lr: 0.000364  loss: 3.4825 (3.6032)  weight_decay: 0.0500 (0.0500)  time: 0.4911  data: 0.1118  max mem: 21002
Epoch: [245]  [150/312]  eta: 0:01:36  lr: 0.000364  min_lr: 0.000364  loss: 3.8376 (3.6142)  weight_decay: 0.0500 (0.0500)  time: 0.4787  data: 0.1041  max mem: 21002
Epoch: [245]  [160/312]  eta: 0:01:29  lr: 0.000363  min_lr: 0.000363  loss: 3.8902 (3.6171)  weight_decay: 0.0500 (0.0500)  time: 0.5650  data: 0.1041  max mem: 21002
Epoch: [245]  [170/312]  eta: 0:01:23  lr: 0.000363  min_lr: 0.000363  loss: 3.8644 (3.6364)  weight_decay: 0.0500 (0.0500)  time: 0.5143  data: 0.1403  max mem: 21002
Epoch: [245]  [180/312]  eta: 0:01:17  lr: 0.000362  min_lr: 0.000362  loss: 3.8654 (3.6435)  weight_decay: 0.0500 (0.0500)  time: 0.5929  data: 0.2358  max mem: 21002
Epoch: [245]  [190/312]  eta: 0:01:09  lr: 0.000362  min_lr: 0.000362  loss: 3.7843 (3.6374)  weight_decay: 0.0500 (0.0500)  time: 0.4561  data: 0.0962  max mem: 21002
Epoch: [245]  [200/312]  eta: 0:01:04  lr: 0.000362  min_lr: 0.000362  loss: 3.6752 (3.6317)  weight_decay: 0.0500 (0.0500)  time: 0.5093  data: 0.1288  max mem: 21002
Epoch: [245]  [210/312]  eta: 0:00:58  lr: 0.000361  min_lr: 0.000361  loss: 3.7685 (3.6388)  weight_decay: 0.0500 (0.0500)  time: 0.6263  data: 0.2023  max mem: 21002
Epoch: [245]  [220/312]  eta: 0:00:51  lr: 0.000361  min_lr: 0.000361  loss: 3.6420 (3.6281)  weight_decay: 0.0500 (0.0500)  time: 0.4061  data: 0.0742  max mem: 21002
Epoch: [245]  [230/312]  eta: 0:00:47  lr: 0.000360  min_lr: 0.000360  loss: 3.5674 (3.6293)  weight_decay: 0.0500 (0.0500)  time: 0.5751  data: 0.1379  max mem: 21002
Epoch: [245]  [240/312]  eta: 0:00:40  lr: 0.000360  min_lr: 0.000360  loss: 3.4795 (3.6160)  weight_decay: 0.0500 (0.0500)  time: 0.6171  data: 0.1379  max mem: 21002
Epoch: [245]  [250/312]  eta: 0:00:35  lr: 0.000359  min_lr: 0.000359  loss: 3.5744 (3.6213)  weight_decay: 0.0500 (0.0500)  time: 0.4265  data: 0.0982  max mem: 21002
Epoch: [245]  [260/312]  eta: 0:00:29  lr: 0.000359  min_lr: 0.000359  loss: 3.8369 (3.6270)  weight_decay: 0.0500 (0.0500)  time: 0.6175  data: 0.2317  max mem: 21002
Epoch: [245]  [270/312]  eta: 0:00:23  lr: 0.000359  min_lr: 0.000359  loss: 3.9268 (3.6235)  weight_decay: 0.0500 (0.0500)  time: 0.5214  data: 0.1343  max mem: 21002
Epoch: [245]  [280/312]  eta: 0:00:18  lr: 0.000358  min_lr: 0.000358  loss: 3.7056 (3.6190)  weight_decay: 0.0500 (0.0500)  time: 0.4623  data: 0.1172  max mem: 21002
Epoch: [245]  [290/312]  eta: 0:00:12  lr: 0.000358  min_lr: 0.000358  loss: 3.8398 (3.6254)  weight_decay: 0.0500 (0.0500)  time: 0.5848  data: 0.1953  max mem: 21002
Epoch: [245]  [300/312]  eta: 0:00:06  lr: 0.000357  min_lr: 0.000357  loss: 3.7663 (3.6251)  weight_decay: 0.0500 (0.0500)  time: 0.4072  data: 0.0786  max mem: 21002
Epoch: [245]  [310/312]  eta: 0:00:01  lr: 0.000357  min_lr: 0.000357  loss: 3.7459 (3.6225)  weight_decay: 0.0500 (0.0500)  time: 0.2772  data: 0.0001  max mem: 21002
Epoch: [245]  [311/312]  eta: 0:00:00  lr: 0.000357  min_lr: 0.000357  loss: 3.7663 (3.6235)  weight_decay: 0.0500 (0.0500)  time: 0.2771  data: 0.0001  max mem: 21002
Epoch: [245] Total time: 0:02:50 (0.5478 s / it)
Averaged stats: lr: 0.000357  min_lr: 0.000357  loss: 3.7663 (3.6531)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.1163 (1.1163)  acc1: 79.1667 (79.1667)  acc5: 94.6615 (94.6615)  time: 8.2577  data: 8.1389  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4108 (1.3598)  acc1: 70.7031 (71.6640)  acc5: 91.1458 (90.8960)  time: 1.0078  data: 0.9112  max mem: 21002
Test: Total time: 0:00:09 (1.0182 s / it)
* Acc@1 71.648 Acc@5 90.912 loss 1.366
Accuracy of the model on the 50000 test images: 71.6%
Max accuracy: 71.65%
Epoch: [246]  [  0/312]  eta: 0:57:45  lr: 0.000357  min_lr: 0.000357  loss: 3.9860 (3.9860)  weight_decay: 0.0500 (0.0500)  time: 11.1090  data: 9.9342  max mem: 21002
Epoch: [246]  [ 10/312]  eta: 0:07:31  lr: 0.000356  min_lr: 0.000356  loss: 3.2744 (3.3755)  weight_decay: 0.0500 (0.0500)  time: 1.4949  data: 1.0187  max mem: 21002
Epoch: [246]  [ 20/312]  eta: 0:04:56  lr: 0.000356  min_lr: 0.000356  loss: 3.3919 (3.4762)  weight_decay: 0.0500 (0.0500)  time: 0.5111  data: 0.1653  max mem: 21002
Epoch: [246]  [ 30/312]  eta: 0:03:40  lr: 0.000356  min_lr: 0.000356  loss: 3.7866 (3.5599)  weight_decay: 0.0500 (0.0500)  time: 0.3879  data: 0.1020  max mem: 21002
Epoch: [246]  [ 40/312]  eta: 0:03:19  lr: 0.000355  min_lr: 0.000355  loss: 3.8029 (3.6335)  weight_decay: 0.0500 (0.0500)  time: 0.4333  data: 0.1466  max mem: 21002
Epoch: [246]  [ 50/312]  eta: 0:03:06  lr: 0.000355  min_lr: 0.000355  loss: 3.9294 (3.6668)  weight_decay: 0.0500 (0.0500)  time: 0.6062  data: 0.3177  max mem: 21002
Epoch: [246]  [ 60/312]  eta: 0:02:43  lr: 0.000354  min_lr: 0.000354  loss: 3.7767 (3.6316)  weight_decay: 0.0500 (0.0500)  time: 0.4768  data: 0.1717  max mem: 21002
Epoch: [246]  [ 70/312]  eta: 0:02:37  lr: 0.000354  min_lr: 0.000354  loss: 3.6909 (3.6841)  weight_decay: 0.0500 (0.0500)  time: 0.4938  data: 0.1917  max mem: 21002
Epoch: [246]  [ 80/312]  eta: 0:02:28  lr: 0.000354  min_lr: 0.000354  loss: 3.9783 (3.7138)  weight_decay: 0.0500 (0.0500)  time: 0.6188  data: 0.3352  max mem: 21002
Epoch: [246]  [ 90/312]  eta: 0:02:15  lr: 0.000353  min_lr: 0.000353  loss: 3.8721 (3.6931)  weight_decay: 0.0500 (0.0500)  time: 0.4714  data: 0.1871  max mem: 21002
Epoch: [246]  [100/312]  eta: 0:02:13  lr: 0.000353  min_lr: 0.000353  loss: 3.4565 (3.6477)  weight_decay: 0.0500 (0.0500)  time: 0.5861  data: 0.2416  max mem: 21002
Epoch: [246]  [110/312]  eta: 0:02:01  lr: 0.000352  min_lr: 0.000352  loss: 3.3311 (3.6338)  weight_decay: 0.0500 (0.0500)  time: 0.5466  data: 0.2024  max mem: 21002
Epoch: [246]  [120/312]  eta: 0:01:55  lr: 0.000352  min_lr: 0.000352  loss: 3.8221 (3.6386)  weight_decay: 0.0500 (0.0500)  time: 0.4583  data: 0.1303  max mem: 21002
Epoch: [246]  [130/312]  eta: 0:01:49  lr: 0.000352  min_lr: 0.000352  loss: 3.8546 (3.6305)  weight_decay: 0.0500 (0.0500)  time: 0.6154  data: 0.2718  max mem: 21002
Epoch: [246]  [140/312]  eta: 0:01:41  lr: 0.000351  min_lr: 0.000351  loss: 3.9405 (3.6469)  weight_decay: 0.0500 (0.0500)  time: 0.5114  data: 0.1461  max mem: 21002
Epoch: [246]  [150/312]  eta: 0:01:36  lr: 0.000351  min_lr: 0.000351  loss: 3.9405 (3.6551)  weight_decay: 0.0500 (0.0500)  time: 0.5677  data: 0.1300  max mem: 21002
Epoch: [246]  [160/312]  eta: 0:01:29  lr: 0.000350  min_lr: 0.000350  loss: 3.9137 (3.6596)  weight_decay: 0.0500 (0.0500)  time: 0.5962  data: 0.2219  max mem: 21002
Epoch: [246]  [170/312]  eta: 0:01:23  lr: 0.000350  min_lr: 0.000350  loss: 3.8436 (3.6638)  weight_decay: 0.0500 (0.0500)  time: 0.5089  data: 0.0976  max mem: 21002
Epoch: [246]  [180/312]  eta: 0:01:16  lr: 0.000350  min_lr: 0.000350  loss: 3.8574 (3.6667)  weight_decay: 0.0500 (0.0500)  time: 0.5076  data: 0.0657  max mem: 21002
Epoch: [246]  [190/312]  eta: 0:01:10  lr: 0.000349  min_lr: 0.000349  loss: 3.8285 (3.6792)  weight_decay: 0.0500 (0.0500)  time: 0.4866  data: 0.0608  max mem: 21002
Epoch: [246]  [200/312]  eta: 0:01:05  lr: 0.000349  min_lr: 0.000349  loss: 3.8285 (3.6911)  weight_decay: 0.0500 (0.0500)  time: 0.6003  data: 0.0902  max mem: 21002
Epoch: [246]  [210/312]  eta: 0:00:59  lr: 0.000348  min_lr: 0.000348  loss: 3.8342 (3.6885)  weight_decay: 0.0500 (0.0500)  time: 0.6039  data: 0.1417  max mem: 21002
Epoch: [246]  [220/312]  eta: 0:00:52  lr: 0.000348  min_lr: 0.000348  loss: 3.7236 (3.6764)  weight_decay: 0.0500 (0.0500)  time: 0.5041  data: 0.0524  max mem: 21002
Epoch: [246]  [230/312]  eta: 0:00:47  lr: 0.000348  min_lr: 0.000348  loss: 3.6543 (3.6712)  weight_decay: 0.0500 (0.0500)  time: 0.5849  data: 0.0286  max mem: 21002
Epoch: [246]  [240/312]  eta: 0:00:40  lr: 0.000347  min_lr: 0.000347  loss: 3.4834 (3.6629)  weight_decay: 0.0500 (0.0500)  time: 0.4841  data: 0.0285  max mem: 21002
Epoch: [246]  [250/312]  eta: 0:00:35  lr: 0.000347  min_lr: 0.000347  loss: 3.2644 (3.6510)  weight_decay: 0.0500 (0.0500)  time: 0.5198  data: 0.0184  max mem: 21002
Epoch: [246]  [260/312]  eta: 0:00:29  lr: 0.000346  min_lr: 0.000346  loss: 3.7561 (3.6549)  weight_decay: 0.0500 (0.0500)  time: 0.5556  data: 0.0351  max mem: 21002
Epoch: [246]  [270/312]  eta: 0:00:23  lr: 0.000346  min_lr: 0.000346  loss: 3.7855 (3.6563)  weight_decay: 0.0500 (0.0500)  time: 0.4785  data: 0.0173  max mem: 21002
Epoch: [246]  [280/312]  eta: 0:00:18  lr: 0.000346  min_lr: 0.000346  loss: 3.7857 (3.6582)  weight_decay: 0.0500 (0.0500)  time: 0.6758  data: 0.0037  max mem: 21002
Epoch: [246]  [290/312]  eta: 0:00:12  lr: 0.000345  min_lr: 0.000345  loss: 3.7161 (3.6541)  weight_decay: 0.0500 (0.0500)  time: 0.5212  data: 0.0035  max mem: 21002
Epoch: [246]  [300/312]  eta: 0:00:06  lr: 0.000345  min_lr: 0.000345  loss: 3.4456 (3.6460)  weight_decay: 0.0500 (0.0500)  time: 0.3187  data: 0.0002  max mem: 21002
Epoch: [246]  [310/312]  eta: 0:00:01  lr: 0.000344  min_lr: 0.000344  loss: 3.6478 (3.6499)  weight_decay: 0.0500 (0.0500)  time: 0.3169  data: 0.0001  max mem: 21002
Epoch: [246]  [311/312]  eta: 0:00:00  lr: 0.000344  min_lr: 0.000344  loss: 3.7191 (3.6504)  weight_decay: 0.0500 (0.0500)  time: 0.3167  data: 0.0001  max mem: 21002
Epoch: [246] Total time: 0:02:52 (0.5515 s / it)
Averaged stats: lr: 0.000344  min_lr: 0.000344  loss: 3.7191 (3.6609)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.0835 (1.0835)  acc1: 80.3385 (80.3385)  acc5: 94.6615 (94.6615)  time: 8.5946  data: 8.4757  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4416 (1.3713)  acc1: 70.5729 (71.9840)  acc5: 91.0156 (91.0880)  time: 1.0753  data: 0.9768  max mem: 21002
Test: Total time: 0:00:09 (1.0847 s / it)
* Acc@1 71.640 Acc@5 90.936 loss 1.381
Accuracy of the model on the 50000 test images: 71.6%
Max accuracy: 71.65%
Epoch: [247]  [  0/312]  eta: 1:01:18  lr: 0.000344  min_lr: 0.000344  loss: 4.1294 (4.1294)  weight_decay: 0.0500 (0.0500)  time: 11.7890  data: 11.4943  max mem: 21002
Epoch: [247]  [ 10/312]  eta: 0:08:01  lr: 0.000344  min_lr: 0.000344  loss: 3.8434 (3.7379)  weight_decay: 0.0500 (0.0500)  time: 1.5930  data: 1.0455  max mem: 21002
Epoch: [247]  [ 20/312]  eta: 0:04:55  lr: 0.000343  min_lr: 0.000343  loss: 3.8086 (3.7234)  weight_decay: 0.0500 (0.0500)  time: 0.4739  data: 0.0406  max mem: 21002
Epoch: [247]  [ 30/312]  eta: 0:03:39  lr: 0.000343  min_lr: 0.000343  loss: 3.5349 (3.6062)  weight_decay: 0.0500 (0.0500)  time: 0.3318  data: 0.0405  max mem: 21002
Epoch: [247]  [ 40/312]  eta: 0:03:12  lr: 0.000343  min_lr: 0.000343  loss: 3.4700 (3.5797)  weight_decay: 0.0500 (0.0500)  time: 0.3892  data: 0.0867  max mem: 21002
Epoch: [247]  [ 50/312]  eta: 0:03:03  lr: 0.000342  min_lr: 0.000342  loss: 3.4664 (3.5129)  weight_decay: 0.0500 (0.0500)  time: 0.5825  data: 0.2802  max mem: 21002
Epoch: [247]  [ 60/312]  eta: 0:02:39  lr: 0.000342  min_lr: 0.000342  loss: 3.4664 (3.5496)  weight_decay: 0.0500 (0.0500)  time: 0.4833  data: 0.1942  max mem: 21002
Epoch: [247]  [ 70/312]  eta: 0:02:34  lr: 0.000341  min_lr: 0.000341  loss: 3.7873 (3.5726)  weight_decay: 0.0500 (0.0500)  time: 0.4739  data: 0.1805  max mem: 21002
Epoch: [247]  [ 80/312]  eta: 0:02:26  lr: 0.000341  min_lr: 0.000341  loss: 3.9008 (3.6345)  weight_decay: 0.0500 (0.0500)  time: 0.6188  data: 0.3271  max mem: 21002
Epoch: [247]  [ 90/312]  eta: 0:02:14  lr: 0.000341  min_lr: 0.000341  loss: 3.8853 (3.6270)  weight_decay: 0.0500 (0.0500)  time: 0.4964  data: 0.1475  max mem: 21002
Epoch: [247]  [100/312]  eta: 0:02:08  lr: 0.000340  min_lr: 0.000340  loss: 3.7209 (3.6426)  weight_decay: 0.0500 (0.0500)  time: 0.5180  data: 0.1294  max mem: 21002
Epoch: [247]  [110/312]  eta: 0:01:57  lr: 0.000340  min_lr: 0.000340  loss: 3.7765 (3.6568)  weight_decay: 0.0500 (0.0500)  time: 0.4546  data: 0.1293  max mem: 21002
Epoch: [247]  [120/312]  eta: 0:01:52  lr: 0.000339  min_lr: 0.000339  loss: 3.8436 (3.6542)  weight_decay: 0.0500 (0.0500)  time: 0.4736  data: 0.1708  max mem: 21002
Epoch: [247]  [130/312]  eta: 0:01:48  lr: 0.000339  min_lr: 0.000339  loss: 3.8436 (3.6566)  weight_decay: 0.0500 (0.0500)  time: 0.6736  data: 0.3457  max mem: 21002
Epoch: [247]  [140/312]  eta: 0:01:38  lr: 0.000339  min_lr: 0.000339  loss: 3.9437 (3.6705)  weight_decay: 0.0500 (0.0500)  time: 0.4901  data: 0.1796  max mem: 21002
Epoch: [247]  [150/312]  eta: 0:01:33  lr: 0.000338  min_lr: 0.000338  loss: 3.9009 (3.6699)  weight_decay: 0.0500 (0.0500)  time: 0.4486  data: 0.1551  max mem: 21002
Epoch: [247]  [160/312]  eta: 0:01:28  lr: 0.000338  min_lr: 0.000338  loss: 3.7086 (3.6623)  weight_decay: 0.0500 (0.0500)  time: 0.6591  data: 0.3536  max mem: 21002
Epoch: [247]  [170/312]  eta: 0:01:20  lr: 0.000337  min_lr: 0.000337  loss: 3.7086 (3.6533)  weight_decay: 0.0500 (0.0500)  time: 0.5011  data: 0.2033  max mem: 21002
Epoch: [247]  [180/312]  eta: 0:01:15  lr: 0.000337  min_lr: 0.000337  loss: 3.7292 (3.6544)  weight_decay: 0.0500 (0.0500)  time: 0.4917  data: 0.2038  max mem: 21002
Epoch: [247]  [190/312]  eta: 0:01:08  lr: 0.000337  min_lr: 0.000337  loss: 3.7282 (3.6596)  weight_decay: 0.0500 (0.0500)  time: 0.4922  data: 0.2048  max mem: 21002
Epoch: [247]  [200/312]  eta: 0:01:03  lr: 0.000336  min_lr: 0.000336  loss: 3.9419 (3.6647)  weight_decay: 0.0500 (0.0500)  time: 0.4944  data: 0.1664  max mem: 21002
Epoch: [247]  [210/312]  eta: 0:00:58  lr: 0.000336  min_lr: 0.000336  loss: 3.9266 (3.6664)  weight_decay: 0.0500 (0.0500)  time: 0.7077  data: 0.3230  max mem: 21002
Epoch: [247]  [220/312]  eta: 0:00:51  lr: 0.000335  min_lr: 0.000335  loss: 3.8044 (3.6698)  weight_decay: 0.0500 (0.0500)  time: 0.5002  data: 0.1589  max mem: 21002
Epoch: [247]  [230/312]  eta: 0:00:46  lr: 0.000335  min_lr: 0.000335  loss: 3.8044 (3.6799)  weight_decay: 0.0500 (0.0500)  time: 0.4783  data: 0.1159  max mem: 21002
Epoch: [247]  [240/312]  eta: 0:00:40  lr: 0.000335  min_lr: 0.000335  loss: 3.7957 (3.6730)  weight_decay: 0.0500 (0.0500)  time: 0.5714  data: 0.1721  max mem: 21002
Epoch: [247]  [250/312]  eta: 0:00:34  lr: 0.000334  min_lr: 0.000334  loss: 3.6088 (3.6739)  weight_decay: 0.0500 (0.0500)  time: 0.4581  data: 0.1044  max mem: 21002
Epoch: [247]  [260/312]  eta: 0:00:29  lr: 0.000334  min_lr: 0.000334  loss: 3.6674 (3.6753)  weight_decay: 0.0500 (0.0500)  time: 0.5273  data: 0.1520  max mem: 21002
Epoch: [247]  [270/312]  eta: 0:00:23  lr: 0.000333  min_lr: 0.000333  loss: 3.7244 (3.6769)  weight_decay: 0.0500 (0.0500)  time: 0.5128  data: 0.1052  max mem: 21002
Epoch: [247]  [280/312]  eta: 0:00:17  lr: 0.000333  min_lr: 0.000333  loss: 3.6677 (3.6780)  weight_decay: 0.0500 (0.0500)  time: 0.5012  data: 0.1116  max mem: 21002
Epoch: [247]  [290/312]  eta: 0:00:12  lr: 0.000333  min_lr: 0.000333  loss: 3.8285 (3.6780)  weight_decay: 0.0500 (0.0500)  time: 0.6066  data: 0.1708  max mem: 21002
Epoch: [247]  [300/312]  eta: 0:00:06  lr: 0.000332  min_lr: 0.000332  loss: 3.7949 (3.6728)  weight_decay: 0.0500 (0.0500)  time: 0.4602  data: 0.0687  max mem: 21002
Epoch: [247]  [310/312]  eta: 0:00:01  lr: 0.000332  min_lr: 0.000332  loss: 3.4360 (3.6656)  weight_decay: 0.0500 (0.0500)  time: 0.2964  data: 0.0131  max mem: 21002
Epoch: [247]  [311/312]  eta: 0:00:00  lr: 0.000332  min_lr: 0.000332  loss: 3.4360 (3.6631)  weight_decay: 0.0500 (0.0500)  time: 0.2880  data: 0.0041  max mem: 21002
Epoch: [247] Total time: 0:02:49 (0.5424 s / it)
Averaged stats: lr: 0.000332  min_lr: 0.000332  loss: 3.4360 (3.6291)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.2156 (1.2156)  acc1: 78.6458 (78.6458)  acc5: 94.0104 (94.0104)  time: 8.8155  data: 8.6966  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4310 (1.3571)  acc1: 71.6146 (71.9680)  acc5: 91.0156 (90.9120)  time: 1.0639  data: 0.9664  max mem: 21002
Test: Total time: 0:00:09 (1.0779 s / it)
* Acc@1 71.536 Acc@5 90.830 loss 1.366
Accuracy of the model on the 50000 test images: 71.5%
Max accuracy: 71.65%
Epoch: [248]  [  0/312]  eta: 1:02:14  lr: 0.000332  min_lr: 0.000332  loss: 3.0823 (3.0823)  weight_decay: 0.0500 (0.0500)  time: 11.9704  data: 8.5388  max mem: 21002
Epoch: [248]  [ 10/312]  eta: 0:07:57  lr: 0.000331  min_lr: 0.000331  loss: 3.7239 (3.5559)  weight_decay: 0.0500 (0.0500)  time: 1.5815  data: 1.0134  max mem: 21002
Epoch: [248]  [ 20/312]  eta: 0:05:16  lr: 0.000331  min_lr: 0.000331  loss: 3.7239 (3.5644)  weight_decay: 0.0500 (0.0500)  time: 0.5394  data: 0.1567  max mem: 21002
Epoch: [248]  [ 30/312]  eta: 0:03:53  lr: 0.000331  min_lr: 0.000331  loss: 3.6366 (3.5900)  weight_decay: 0.0500 (0.0500)  time: 0.4137  data: 0.0266  max mem: 21002
Epoch: [248]  [ 40/312]  eta: 0:03:19  lr: 0.000330  min_lr: 0.000330  loss: 3.8773 (3.6103)  weight_decay: 0.0500 (0.0500)  time: 0.3655  data: 0.0322  max mem: 21002
Epoch: [248]  [ 50/312]  eta: 0:03:14  lr: 0.000330  min_lr: 0.000330  loss: 3.8773 (3.6557)  weight_decay: 0.0500 (0.0500)  time: 0.6122  data: 0.1746  max mem: 21002
Epoch: [248]  [ 60/312]  eta: 0:02:48  lr: 0.000329  min_lr: 0.000329  loss: 3.5915 (3.5980)  weight_decay: 0.0500 (0.0500)  time: 0.5348  data: 0.1432  max mem: 21002
Epoch: [248]  [ 70/312]  eta: 0:02:40  lr: 0.000329  min_lr: 0.000329  loss: 3.5329 (3.6424)  weight_decay: 0.0500 (0.0500)  time: 0.4599  data: 0.1288  max mem: 21002
Epoch: [248]  [ 80/312]  eta: 0:02:29  lr: 0.000329  min_lr: 0.000329  loss: 3.8326 (3.6461)  weight_decay: 0.0500 (0.0500)  time: 0.5779  data: 0.1771  max mem: 21002
Epoch: [248]  [ 90/312]  eta: 0:02:18  lr: 0.000328  min_lr: 0.000328  loss: 3.6951 (3.6431)  weight_decay: 0.0500 (0.0500)  time: 0.4813  data: 0.1289  max mem: 21002
Epoch: [248]  [100/312]  eta: 0:02:13  lr: 0.000328  min_lr: 0.000328  loss: 3.6788 (3.6321)  weight_decay: 0.0500 (0.0500)  time: 0.5560  data: 0.1987  max mem: 21002
Epoch: [248]  [110/312]  eta: 0:02:00  lr: 0.000327  min_lr: 0.000327  loss: 3.8739 (3.6574)  weight_decay: 0.0500 (0.0500)  time: 0.4795  data: 0.1188  max mem: 21002
Epoch: [248]  [120/312]  eta: 0:01:55  lr: 0.000327  min_lr: 0.000327  loss: 3.8160 (3.6387)  weight_decay: 0.0500 (0.0500)  time: 0.4749  data: 0.1293  max mem: 21002
Epoch: [248]  [130/312]  eta: 0:01:50  lr: 0.000327  min_lr: 0.000327  loss: 3.4364 (3.6342)  weight_decay: 0.0500 (0.0500)  time: 0.6545  data: 0.2660  max mem: 21002
Epoch: [248]  [140/312]  eta: 0:01:40  lr: 0.000326  min_lr: 0.000326  loss: 3.3945 (3.6163)  weight_decay: 0.0500 (0.0500)  time: 0.4683  data: 0.1374  max mem: 21002
Epoch: [248]  [150/312]  eta: 0:01:35  lr: 0.000326  min_lr: 0.000326  loss: 3.6748 (3.6163)  weight_decay: 0.0500 (0.0500)  time: 0.4638  data: 0.1482  max mem: 21002
Epoch: [248]  [160/312]  eta: 0:01:29  lr: 0.000325  min_lr: 0.000325  loss: 3.5871 (3.6046)  weight_decay: 0.0500 (0.0500)  time: 0.6163  data: 0.2408  max mem: 21002
Epoch: [248]  [170/312]  eta: 0:01:21  lr: 0.000325  min_lr: 0.000325  loss: 3.5596 (3.5973)  weight_decay: 0.0500 (0.0500)  time: 0.4918  data: 0.1442  max mem: 21002
Epoch: [248]  [180/312]  eta: 0:01:16  lr: 0.000325  min_lr: 0.000325  loss: 3.8662 (3.6043)  weight_decay: 0.0500 (0.0500)  time: 0.4938  data: 0.2029  max mem: 21002
Epoch: [248]  [190/312]  eta: 0:01:08  lr: 0.000324  min_lr: 0.000324  loss: 3.7909 (3.6035)  weight_decay: 0.0500 (0.0500)  time: 0.4469  data: 0.1520  max mem: 21002
Epoch: [248]  [200/312]  eta: 0:01:03  lr: 0.000324  min_lr: 0.000324  loss: 3.5846 (3.5979)  weight_decay: 0.0500 (0.0500)  time: 0.4879  data: 0.1958  max mem: 21002
Epoch: [248]  [210/312]  eta: 0:00:58  lr: 0.000323  min_lr: 0.000323  loss: 3.6686 (3.6087)  weight_decay: 0.0500 (0.0500)  time: 0.6815  data: 0.3934  max mem: 21002
Epoch: [248]  [220/312]  eta: 0:00:51  lr: 0.000323  min_lr: 0.000323  loss: 3.6686 (3.6070)  weight_decay: 0.0500 (0.0500)  time: 0.4878  data: 0.1981  max mem: 21002
Epoch: [248]  [230/312]  eta: 0:00:46  lr: 0.000323  min_lr: 0.000323  loss: 3.8928 (3.6202)  weight_decay: 0.0500 (0.0500)  time: 0.5051  data: 0.1788  max mem: 21002
Epoch: [248]  [240/312]  eta: 0:00:40  lr: 0.000322  min_lr: 0.000322  loss: 3.9625 (3.6268)  weight_decay: 0.0500 (0.0500)  time: 0.5951  data: 0.2454  max mem: 21002
Epoch: [248]  [250/312]  eta: 0:00:34  lr: 0.000322  min_lr: 0.000322  loss: 3.6850 (3.6263)  weight_decay: 0.0500 (0.0500)  time: 0.5186  data: 0.1411  max mem: 21002
Epoch: [248]  [260/312]  eta: 0:00:29  lr: 0.000322  min_lr: 0.000322  loss: 3.6850 (3.6308)  weight_decay: 0.0500 (0.0500)  time: 0.5882  data: 0.1836  max mem: 21002
Epoch: [248]  [270/312]  eta: 0:00:23  lr: 0.000321  min_lr: 0.000321  loss: 3.6827 (3.6337)  weight_decay: 0.0500 (0.0500)  time: 0.4474  data: 0.1101  max mem: 21002
Epoch: [248]  [280/312]  eta: 0:00:17  lr: 0.000321  min_lr: 0.000321  loss: 3.6938 (3.6396)  weight_decay: 0.0500 (0.0500)  time: 0.4534  data: 0.1524  max mem: 21002
Epoch: [248]  [290/312]  eta: 0:00:12  lr: 0.000320  min_lr: 0.000320  loss: 3.6938 (3.6323)  weight_decay: 0.0500 (0.0500)  time: 0.6234  data: 0.2913  max mem: 21002
Epoch: [248]  [300/312]  eta: 0:00:06  lr: 0.000320  min_lr: 0.000320  loss: 3.7362 (3.6372)  weight_decay: 0.0500 (0.0500)  time: 0.4508  data: 0.1396  max mem: 21002
Epoch: [248]  [310/312]  eta: 0:00:01  lr: 0.000320  min_lr: 0.000320  loss: 3.7968 (3.6363)  weight_decay: 0.0500 (0.0500)  time: 0.2775  data: 0.0001  max mem: 21002
Epoch: [248]  [311/312]  eta: 0:00:00  lr: 0.000320  min_lr: 0.000320  loss: 3.7771 (3.6352)  weight_decay: 0.0500 (0.0500)  time: 0.2774  data: 0.0001  max mem: 21002
Epoch: [248] Total time: 0:02:49 (0.5448 s / it)
Averaged stats: lr: 0.000320  min_lr: 0.000320  loss: 3.7771 (3.6417)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.1173 (1.1173)  acc1: 79.2969 (79.2969)  acc5: 94.4010 (94.4010)  time: 8.3511  data: 8.2328  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4428 (1.3355)  acc1: 70.5729 (71.8880)  acc5: 90.3646 (90.9280)  time: 1.0215  data: 0.9239  max mem: 21002
Test: Total time: 0:00:09 (1.0314 s / it)
* Acc@1 71.804 Acc@5 90.836 loss 1.349
Accuracy of the model on the 50000 test images: 71.8%
Max accuracy: 71.80%
Epoch: [249]  [  0/312]  eta: 1:00:13  lr: 0.000320  min_lr: 0.000320  loss: 3.5886 (3.5886)  weight_decay: 0.0500 (0.0500)  time: 11.5820  data: 9.0084  max mem: 21002
Epoch: [249]  [ 10/312]  eta: 0:07:54  lr: 0.000319  min_lr: 0.000319  loss: 3.5886 (3.5954)  weight_decay: 0.0500 (0.0500)  time: 1.5714  data: 1.0167  max mem: 21002
Epoch: [249]  [ 20/312]  eta: 0:05:15  lr: 0.000319  min_lr: 0.000319  loss: 3.5422 (3.5266)  weight_decay: 0.0500 (0.0500)  time: 0.5537  data: 0.1464  max mem: 21002
Epoch: [249]  [ 30/312]  eta: 0:03:52  lr: 0.000318  min_lr: 0.000318  loss: 3.5365 (3.5457)  weight_decay: 0.0500 (0.0500)  time: 0.4135  data: 0.0380  max mem: 21002
Epoch: [249]  [ 40/312]  eta: 0:03:10  lr: 0.000318  min_lr: 0.000318  loss: 3.5400 (3.5551)  weight_decay: 0.0500 (0.0500)  time: 0.3047  data: 0.0007  max mem: 21002
Epoch: [249]  [ 50/312]  eta: 0:03:02  lr: 0.000318  min_lr: 0.000318  loss: 3.8488 (3.6038)  weight_decay: 0.0500 (0.0500)  time: 0.4970  data: 0.1763  max mem: 21002
Epoch: [249]  [ 60/312]  eta: 0:02:38  lr: 0.000317  min_lr: 0.000317  loss: 3.8568 (3.5991)  weight_decay: 0.0500 (0.0500)  time: 0.4799  data: 0.1764  max mem: 21002
Epoch: [249]  [ 70/312]  eta: 0:02:34  lr: 0.000317  min_lr: 0.000317  loss: 3.8568 (3.6509)  weight_decay: 0.0500 (0.0500)  time: 0.4877  data: 0.1739  max mem: 21002
Epoch: [249]  [ 80/312]  eta: 0:02:25  lr: 0.000316  min_lr: 0.000316  loss: 3.8754 (3.6697)  weight_decay: 0.0500 (0.0500)  time: 0.6245  data: 0.2534  max mem: 21002
Epoch: [249]  [ 90/312]  eta: 0:02:16  lr: 0.000316  min_lr: 0.000316  loss: 3.8551 (3.6589)  weight_decay: 0.0500 (0.0500)  time: 0.5265  data: 0.1433  max mem: 21002
Epoch: [249]  [100/312]  eta: 0:02:09  lr: 0.000316  min_lr: 0.000316  loss: 3.8379 (3.6676)  weight_decay: 0.0500 (0.0500)  time: 0.5516  data: 0.1660  max mem: 21002
Epoch: [249]  [110/312]  eta: 0:01:57  lr: 0.000315  min_lr: 0.000315  loss: 3.8379 (3.6902)  weight_decay: 0.0500 (0.0500)  time: 0.4476  data: 0.1029  max mem: 21002
Epoch: [249]  [120/312]  eta: 0:01:52  lr: 0.000315  min_lr: 0.000315  loss: 3.6917 (3.6781)  weight_decay: 0.0500 (0.0500)  time: 0.4545  data: 0.0937  max mem: 21002
Epoch: [249]  [130/312]  eta: 0:01:48  lr: 0.000314  min_lr: 0.000314  loss: 3.8176 (3.7038)  weight_decay: 0.0500 (0.0500)  time: 0.6550  data: 0.1742  max mem: 21002
Epoch: [249]  [140/312]  eta: 0:01:39  lr: 0.000314  min_lr: 0.000314  loss: 3.9176 (3.6973)  weight_decay: 0.0500 (0.0500)  time: 0.5188  data: 0.1152  max mem: 21002
Epoch: [249]  [150/312]  eta: 0:01:34  lr: 0.000314  min_lr: 0.000314  loss: 3.4525 (3.6734)  weight_decay: 0.0500 (0.0500)  time: 0.5213  data: 0.0951  max mem: 21002
Epoch: [249]  [160/312]  eta: 0:01:27  lr: 0.000313  min_lr: 0.000313  loss: 3.4452 (3.6655)  weight_decay: 0.0500 (0.0500)  time: 0.5662  data: 0.0680  max mem: 21002
Epoch: [249]  [170/312]  eta: 0:01:21  lr: 0.000313  min_lr: 0.000313  loss: 3.5527 (3.6603)  weight_decay: 0.0500 (0.0500)  time: 0.4774  data: 0.0648  max mem: 21002
Epoch: [249]  [180/312]  eta: 0:01:16  lr: 0.000313  min_lr: 0.000313  loss: 3.5527 (3.6563)  weight_decay: 0.0500 (0.0500)  time: 0.5900  data: 0.1094  max mem: 21002
Epoch: [249]  [190/312]  eta: 0:01:08  lr: 0.000312  min_lr: 0.000312  loss: 3.6545 (3.6606)  weight_decay: 0.0500 (0.0500)  time: 0.4806  data: 0.0565  max mem: 21002
Epoch: [249]  [200/312]  eta: 0:01:03  lr: 0.000312  min_lr: 0.000312  loss: 3.6545 (3.6508)  weight_decay: 0.0500 (0.0500)  time: 0.4863  data: 0.0630  max mem: 21002
Epoch: [249]  [210/312]  eta: 0:00:58  lr: 0.000311  min_lr: 0.000311  loss: 3.5850 (3.6453)  weight_decay: 0.0500 (0.0500)  time: 0.6512  data: 0.1003  max mem: 21002
Epoch: [249]  [220/312]  eta: 0:00:51  lr: 0.000311  min_lr: 0.000311  loss: 3.4050 (3.6368)  weight_decay: 0.0500 (0.0500)  time: 0.4854  data: 0.0718  max mem: 21002
Epoch: [249]  [230/312]  eta: 0:00:46  lr: 0.000311  min_lr: 0.000311  loss: 3.4264 (3.6373)  weight_decay: 0.0500 (0.0500)  time: 0.5113  data: 0.0836  max mem: 21002
Epoch: [249]  [240/312]  eta: 0:00:40  lr: 0.000310  min_lr: 0.000310  loss: 3.7432 (3.6324)  weight_decay: 0.0500 (0.0500)  time: 0.6301  data: 0.0542  max mem: 21002
Epoch: [249]  [250/312]  eta: 0:00:34  lr: 0.000310  min_lr: 0.000310  loss: 3.7432 (3.6322)  weight_decay: 0.0500 (0.0500)  time: 0.4754  data: 0.0346  max mem: 21002
Epoch: [249]  [260/312]  eta: 0:00:29  lr: 0.000309  min_lr: 0.000309  loss: 3.7240 (3.6321)  weight_decay: 0.0500 (0.0500)  time: 0.5223  data: 0.0504  max mem: 21002
Epoch: [249]  [270/312]  eta: 0:00:23  lr: 0.000309  min_lr: 0.000309  loss: 3.6738 (3.6301)  weight_decay: 0.0500 (0.0500)  time: 0.4964  data: 0.0314  max mem: 21002
Epoch: [249]  [280/312]  eta: 0:00:17  lr: 0.000309  min_lr: 0.000309  loss: 3.8097 (3.6352)  weight_decay: 0.0500 (0.0500)  time: 0.5108  data: 0.0550  max mem: 21002
Epoch: [249]  [290/312]  eta: 0:00:12  lr: 0.000308  min_lr: 0.000308  loss: 3.7033 (3.6278)  weight_decay: 0.0500 (0.0500)  time: 0.6335  data: 0.0505  max mem: 21002
Epoch: [249]  [300/312]  eta: 0:00:06  lr: 0.000308  min_lr: 0.000308  loss: 3.3602 (3.6261)  weight_decay: 0.0500 (0.0500)  time: 0.4431  data: 0.0351  max mem: 21002
Epoch: [249]  [310/312]  eta: 0:00:01  lr: 0.000308  min_lr: 0.000308  loss: 3.7005 (3.6255)  weight_decay: 0.0500 (0.0500)  time: 0.3004  data: 0.0244  max mem: 21002
Epoch: [249]  [311/312]  eta: 0:00:00  lr: 0.000308  min_lr: 0.000308  loss: 3.7381 (3.6271)  weight_decay: 0.0500 (0.0500)  time: 0.2926  data: 0.0160  max mem: 21002
Epoch: [249] Total time: 0:02:50 (0.5454 s / it)
Averaged stats: lr: 0.000308  min_lr: 0.000308  loss: 3.7381 (3.6271)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.1653 (1.1653)  acc1: 78.2552 (78.2552)  acc5: 94.6615 (94.6615)  time: 8.7596  data: 8.6409  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4441 (1.4043)  acc1: 71.8750 (71.9200)  acc5: 91.2760 (91.3440)  time: 1.0589  data: 0.9602  max mem: 21002
Test: Total time: 0:00:09 (1.0858 s / it)
* Acc@1 71.632 Acc@5 90.964 loss 1.414
Accuracy of the model on the 50000 test images: 71.6%
Max accuracy: 71.80%
Epoch: [250]  [  0/312]  eta: 1:03:37  lr: 0.000307  min_lr: 0.000307  loss: 4.1180 (4.1180)  weight_decay: 0.0500 (0.0500)  time: 12.2350  data: 8.5905  max mem: 21002
Epoch: [250]  [ 10/312]  eta: 0:08:13  lr: 0.000307  min_lr: 0.000307  loss: 4.0028 (3.7599)  weight_decay: 0.0500 (0.0500)  time: 1.6337  data: 0.9893  max mem: 21002
Epoch: [250]  [ 20/312]  eta: 0:05:23  lr: 0.000307  min_lr: 0.000307  loss: 3.8929 (3.7260)  weight_decay: 0.0500 (0.0500)  time: 0.5500  data: 0.1660  max mem: 21002
Epoch: [250]  [ 30/312]  eta: 0:03:57  lr: 0.000306  min_lr: 0.000306  loss: 3.7333 (3.6636)  weight_decay: 0.0500 (0.0500)  time: 0.4095  data: 0.0533  max mem: 21002
Epoch: [250]  [ 40/312]  eta: 0:03:20  lr: 0.000306  min_lr: 0.000306  loss: 3.6906 (3.6522)  weight_decay: 0.0500 (0.0500)  time: 0.3507  data: 0.0024  max mem: 21002
Epoch: [250]  [ 50/312]  eta: 0:03:16  lr: 0.000306  min_lr: 0.000306  loss: 3.6906 (3.6426)  weight_decay: 0.0500 (0.0500)  time: 0.6000  data: 0.0139  max mem: 21002
Epoch: [250]  [ 60/312]  eta: 0:02:54  lr: 0.000305  min_lr: 0.000305  loss: 3.7956 (3.6882)  weight_decay: 0.0500 (0.0500)  time: 0.6034  data: 0.0748  max mem: 21002
Epoch: [250]  [ 70/312]  eta: 0:02:43  lr: 0.000305  min_lr: 0.000305  loss: 3.4195 (3.6219)  weight_decay: 0.0500 (0.0500)  time: 0.4846  data: 0.0618  max mem: 21002
Epoch: [250]  [ 80/312]  eta: 0:02:32  lr: 0.000304  min_lr: 0.000304  loss: 3.1487 (3.5860)  weight_decay: 0.0500 (0.0500)  time: 0.5478  data: 0.0017  max mem: 21002
Epoch: [250]  [ 90/312]  eta: 0:02:20  lr: 0.000304  min_lr: 0.000304  loss: 3.4762 (3.6166)  weight_decay: 0.0500 (0.0500)  time: 0.4844  data: 0.0159  max mem: 21002
Epoch: [250]  [100/312]  eta: 0:02:15  lr: 0.000304  min_lr: 0.000304  loss: 3.8466 (3.6215)  weight_decay: 0.0500 (0.0500)  time: 0.5669  data: 0.0548  max mem: 21002
Epoch: [250]  [110/312]  eta: 0:02:02  lr: 0.000303  min_lr: 0.000303  loss: 3.8466 (3.6335)  weight_decay: 0.0500 (0.0500)  time: 0.4989  data: 0.0405  max mem: 21002
Epoch: [250]  [120/312]  eta: 0:01:57  lr: 0.000303  min_lr: 0.000303  loss: 3.7318 (3.6323)  weight_decay: 0.0500 (0.0500)  time: 0.4843  data: 0.0501  max mem: 21002
Epoch: [250]  [130/312]  eta: 0:01:53  lr: 0.000303  min_lr: 0.000303  loss: 3.4965 (3.6148)  weight_decay: 0.0500 (0.0500)  time: 0.6970  data: 0.0811  max mem: 21002
Epoch: [250]  [140/312]  eta: 0:01:42  lr: 0.000302  min_lr: 0.000302  loss: 3.7549 (3.6222)  weight_decay: 0.0500 (0.0500)  time: 0.5023  data: 0.0326  max mem: 21002
Epoch: [250]  [150/312]  eta: 0:01:38  lr: 0.000302  min_lr: 0.000302  loss: 3.8219 (3.6377)  weight_decay: 0.0500 (0.0500)  time: 0.5027  data: 0.0335  max mem: 21002
Epoch: [250]  [160/312]  eta: 0:01:31  lr: 0.000301  min_lr: 0.000301  loss: 3.7931 (3.6266)  weight_decay: 0.0500 (0.0500)  time: 0.6317  data: 0.0566  max mem: 21002
Epoch: [250]  [170/312]  eta: 0:01:24  lr: 0.000301  min_lr: 0.000301  loss: 3.5204 (3.6204)  weight_decay: 0.0500 (0.0500)  time: 0.4964  data: 0.0413  max mem: 21002
Epoch: [250]  [180/312]  eta: 0:01:19  lr: 0.000301  min_lr: 0.000301  loss: 3.7237 (3.6293)  weight_decay: 0.0500 (0.0500)  time: 0.5796  data: 0.0241  max mem: 21002
Epoch: [250]  [190/312]  eta: 0:01:11  lr: 0.000300  min_lr: 0.000300  loss: 3.7431 (3.6330)  weight_decay: 0.0500 (0.0500)  time: 0.5093  data: 0.0135  max mem: 21002
Epoch: [250]  [200/312]  eta: 0:01:05  lr: 0.000300  min_lr: 0.000300  loss: 3.7869 (3.6407)  weight_decay: 0.0500 (0.0500)  time: 0.4603  data: 0.0193  max mem: 21002
Epoch: [250]  [210/312]  eta: 0:01:00  lr: 0.000299  min_lr: 0.000299  loss: 3.7841 (3.6427)  weight_decay: 0.0500 (0.0500)  time: 0.6388  data: 0.0314  max mem: 21002
Epoch: [250]  [220/312]  eta: 0:00:53  lr: 0.000299  min_lr: 0.000299  loss: 3.5892 (3.6332)  weight_decay: 0.0500 (0.0500)  time: 0.5079  data: 0.0505  max mem: 21002
Epoch: [250]  [230/312]  eta: 0:00:47  lr: 0.000299  min_lr: 0.000299  loss: 3.6645 (3.6366)  weight_decay: 0.0500 (0.0500)  time: 0.4629  data: 0.0768  max mem: 21002
Epoch: [250]  [240/312]  eta: 0:00:41  lr: 0.000298  min_lr: 0.000298  loss: 3.8002 (3.6401)  weight_decay: 0.0500 (0.0500)  time: 0.6061  data: 0.0452  max mem: 21002
Epoch: [250]  [250/312]  eta: 0:00:35  lr: 0.000298  min_lr: 0.000298  loss: 3.8337 (3.6383)  weight_decay: 0.0500 (0.0500)  time: 0.5564  data: 0.0008  max mem: 21002
Epoch: [250]  [260/312]  eta: 0:00:29  lr: 0.000298  min_lr: 0.000298  loss: 3.9192 (3.6508)  weight_decay: 0.0500 (0.0500)  time: 0.4680  data: 0.0571  max mem: 21002
Epoch: [250]  [270/312]  eta: 0:00:23  lr: 0.000297  min_lr: 0.000297  loss: 3.7754 (3.6504)  weight_decay: 0.0500 (0.0500)  time: 0.3769  data: 0.0571  max mem: 21002
Epoch: [250]  [280/312]  eta: 0:00:18  lr: 0.000297  min_lr: 0.000297  loss: 3.4518 (3.6425)  weight_decay: 0.0500 (0.0500)  time: 0.5783  data: 0.0218  max mem: 21002
Epoch: [250]  [290/312]  eta: 0:00:12  lr: 0.000296  min_lr: 0.000296  loss: 3.8858 (3.6528)  weight_decay: 0.0500 (0.0500)  time: 0.6494  data: 0.0325  max mem: 21002
Epoch: [250]  [300/312]  eta: 0:00:06  lr: 0.000296  min_lr: 0.000296  loss: 3.7939 (3.6524)  weight_decay: 0.0500 (0.0500)  time: 0.3554  data: 0.0111  max mem: 21002
Epoch: [250]  [310/312]  eta: 0:00:01  lr: 0.000296  min_lr: 0.000296  loss: 3.7349 (3.6602)  weight_decay: 0.0500 (0.0500)  time: 0.2771  data: 0.0001  max mem: 21002
Epoch: [250]  [311/312]  eta: 0:00:00  lr: 0.000296  min_lr: 0.000296  loss: 3.7349 (3.6576)  weight_decay: 0.0500 (0.0500)  time: 0.2771  data: 0.0001  max mem: 21002
Epoch: [250] Total time: 0:02:52 (0.5524 s / it)
Averaged stats: lr: 0.000296  min_lr: 0.000296  loss: 3.7349 (3.6425)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 1.1918 (1.1918)  acc1: 78.5156 (78.5156)  acc5: 94.6615 (94.6615)  time: 8.9842  data: 8.8655  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4628 (1.3902)  acc1: 70.9635 (72.3520)  acc5: 90.6250 (90.7200)  time: 1.0839  data: 0.9851  max mem: 21002
Test: Total time: 0:00:09 (1.0944 s / it)
* Acc@1 72.080 Acc@5 90.992 loss 1.394
Accuracy of the model on the 50000 test images: 72.1%
Max accuracy: 72.08%
Epoch: [251]  [  0/312]  eta: 1:00:07  lr: 0.000296  min_lr: 0.000296  loss: 3.9908 (3.9908)  weight_decay: 0.0500 (0.0500)  time: 11.5623  data: 10.9183  max mem: 21002
Epoch: [251]  [ 10/312]  eta: 0:08:04  lr: 0.000295  min_lr: 0.000295  loss: 3.9908 (3.8718)  weight_decay: 0.0500 (0.0500)  time: 1.6054  data: 0.9932  max mem: 21002
Epoch: [251]  [ 20/312]  eta: 0:05:01  lr: 0.000295  min_lr: 0.000295  loss: 3.8876 (3.8268)  weight_decay: 0.0500 (0.0500)  time: 0.5060  data: 0.0281  max mem: 21002
Epoch: [251]  [ 30/312]  eta: 0:03:43  lr: 0.000295  min_lr: 0.000295  loss: 3.8249 (3.7891)  weight_decay: 0.0500 (0.0500)  time: 0.3450  data: 0.0281  max mem: 21002
Epoch: [251]  [ 40/312]  eta: 0:03:09  lr: 0.000294  min_lr: 0.000294  loss: 3.7646 (3.7279)  weight_decay: 0.0500 (0.0500)  time: 0.3447  data: 0.0580  max mem: 21002
Epoch: [251]  [ 50/312]  eta: 0:02:59  lr: 0.000294  min_lr: 0.000294  loss: 3.6712 (3.6800)  weight_decay: 0.0500 (0.0500)  time: 0.5202  data: 0.2347  max mem: 21002
Epoch: [251]  [ 60/312]  eta: 0:02:36  lr: 0.000293  min_lr: 0.000293  loss: 3.7713 (3.7088)  weight_decay: 0.0500 (0.0500)  time: 0.4620  data: 0.1774  max mem: 21002
Epoch: [251]  [ 70/312]  eta: 0:02:31  lr: 0.000293  min_lr: 0.000293  loss: 3.8724 (3.7108)  weight_decay: 0.0500 (0.0500)  time: 0.4670  data: 0.1819  max mem: 21002
Epoch: [251]  [ 80/312]  eta: 0:02:24  lr: 0.000293  min_lr: 0.000293  loss: 3.6738 (3.6874)  weight_decay: 0.0500 (0.0500)  time: 0.6398  data: 0.3537  max mem: 21002
Epoch: [251]  [ 90/312]  eta: 0:02:12  lr: 0.000292  min_lr: 0.000292  loss: 3.7612 (3.7003)  weight_decay: 0.0500 (0.0500)  time: 0.4928  data: 0.2058  max mem: 21002
Epoch: [251]  [100/312]  eta: 0:02:08  lr: 0.000292  min_lr: 0.000292  loss: 3.6682 (3.6930)  weight_decay: 0.0500 (0.0500)  time: 0.5290  data: 0.2269  max mem: 21002
Epoch: [251]  [110/312]  eta: 0:01:56  lr: 0.000292  min_lr: 0.000292  loss: 3.6682 (3.6996)  weight_decay: 0.0500 (0.0500)  time: 0.4969  data: 0.1936  max mem: 21002
Epoch: [251]  [120/312]  eta: 0:01:51  lr: 0.000291  min_lr: 0.000291  loss: 3.7862 (3.6920)  weight_decay: 0.0500 (0.0500)  time: 0.4551  data: 0.1648  max mem: 21002
Epoch: [251]  [130/312]  eta: 0:01:46  lr: 0.000291  min_lr: 0.000291  loss: 3.6517 (3.6821)  weight_decay: 0.0500 (0.0500)  time: 0.6361  data: 0.3324  max mem: 21002
Epoch: [251]  [140/312]  eta: 0:01:37  lr: 0.000290  min_lr: 0.000290  loss: 3.6629 (3.6813)  weight_decay: 0.0500 (0.0500)  time: 0.4824  data: 0.1684  max mem: 21002
Epoch: [251]  [150/312]  eta: 0:01:32  lr: 0.000290  min_lr: 0.000290  loss: 3.6228 (3.6692)  weight_decay: 0.0500 (0.0500)  time: 0.4840  data: 0.1861  max mem: 21002
Epoch: [251]  [160/312]  eta: 0:01:27  lr: 0.000290  min_lr: 0.000290  loss: 3.6053 (3.6742)  weight_decay: 0.0500 (0.0500)  time: 0.6431  data: 0.3413  max mem: 21002
Epoch: [251]  [170/312]  eta: 0:01:20  lr: 0.000289  min_lr: 0.000289  loss: 3.8912 (3.6770)  weight_decay: 0.0500 (0.0500)  time: 0.4950  data: 0.1937  max mem: 21002
Epoch: [251]  [180/312]  eta: 0:01:15  lr: 0.000289  min_lr: 0.000289  loss: 3.8558 (3.6779)  weight_decay: 0.0500 (0.0500)  time: 0.5099  data: 0.2001  max mem: 21002
Epoch: [251]  [190/312]  eta: 0:01:07  lr: 0.000289  min_lr: 0.000289  loss: 3.8361 (3.6796)  weight_decay: 0.0500 (0.0500)  time: 0.4733  data: 0.1623  max mem: 21002
Epoch: [251]  [200/312]  eta: 0:01:02  lr: 0.000288  min_lr: 0.000288  loss: 3.9802 (3.6805)  weight_decay: 0.0500 (0.0500)  time: 0.4746  data: 0.1860  max mem: 21002
Epoch: [251]  [210/312]  eta: 0:00:58  lr: 0.000288  min_lr: 0.000288  loss: 3.3604 (3.6580)  weight_decay: 0.0500 (0.0500)  time: 0.7163  data: 0.3786  max mem: 21002
Epoch: [251]  [220/312]  eta: 0:00:51  lr: 0.000287  min_lr: 0.000287  loss: 3.8866 (3.6670)  weight_decay: 0.0500 (0.0500)  time: 0.5284  data: 0.1932  max mem: 21002
Epoch: [251]  [230/312]  eta: 0:00:46  lr: 0.000287  min_lr: 0.000287  loss: 3.9003 (3.6726)  weight_decay: 0.0500 (0.0500)  time: 0.4731  data: 0.1428  max mem: 21002
Epoch: [251]  [240/312]  eta: 0:00:40  lr: 0.000287  min_lr: 0.000287  loss: 3.8252 (3.6730)  weight_decay: 0.0500 (0.0500)  time: 0.6219  data: 0.2262  max mem: 21002
Epoch: [251]  [250/312]  eta: 0:00:34  lr: 0.000286  min_lr: 0.000286  loss: 3.9237 (3.6809)  weight_decay: 0.0500 (0.0500)  time: 0.5681  data: 0.1723  max mem: 21002
Epoch: [251]  [260/312]  eta: 0:00:29  lr: 0.000286  min_lr: 0.000286  loss: 3.6617 (3.6719)  weight_decay: 0.0500 (0.0500)  time: 0.5797  data: 0.1509  max mem: 21002
Epoch: [251]  [270/312]  eta: 0:00:23  lr: 0.000286  min_lr: 0.000286  loss: 3.6617 (3.6662)  weight_decay: 0.0500 (0.0500)  time: 0.4483  data: 0.0626  max mem: 21002
Epoch: [251]  [280/312]  eta: 0:00:17  lr: 0.000285  min_lr: 0.000285  loss: 3.6846 (3.6593)  weight_decay: 0.0500 (0.0500)  time: 0.5146  data: 0.1127  max mem: 21002
Epoch: [251]  [290/312]  eta: 0:00:12  lr: 0.000285  min_lr: 0.000285  loss: 3.6921 (3.6584)  weight_decay: 0.0500 (0.0500)  time: 0.6344  data: 0.1799  max mem: 21002
Epoch: [251]  [300/312]  eta: 0:00:06  lr: 0.000284  min_lr: 0.000284  loss: 3.7091 (3.6638)  weight_decay: 0.0500 (0.0500)  time: 0.4031  data: 0.0676  max mem: 21002
Epoch: [251]  [310/312]  eta: 0:00:01  lr: 0.000284  min_lr: 0.000284  loss: 3.6435 (3.6618)  weight_decay: 0.0500 (0.0500)  time: 0.2774  data: 0.0001  max mem: 21002
Epoch: [251]  [311/312]  eta: 0:00:00  lr: 0.000284  min_lr: 0.000284  loss: 3.6435 (3.6630)  weight_decay: 0.0500 (0.0500)  time: 0.2773  data: 0.0001  max mem: 21002
Epoch: [251] Total time: 0:02:49 (0.5439 s / it)
Averaged stats: lr: 0.000284  min_lr: 0.000284  loss: 3.6435 (3.6336)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:21  loss: 1.1007 (1.1007)  acc1: 80.7292 (80.7292)  acc5: 94.1406 (94.1406)  time: 9.0145  data: 8.8957  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4347 (1.3840)  acc1: 72.0052 (72.1440)  acc5: 91.6667 (91.0400)  time: 1.0889  data: 0.9885  max mem: 21002
Test: Total time: 0:00:10 (1.1275 s / it)
* Acc@1 71.820 Acc@5 90.914 loss 1.397
Accuracy of the model on the 50000 test images: 71.8%
Max accuracy: 72.08%
Epoch: [252]  [  0/312]  eta: 1:03:38  lr: 0.000284  min_lr: 0.000284  loss: 3.3697 (3.3697)  weight_decay: 0.0500 (0.0500)  time: 12.2390  data: 9.2059  max mem: 21002
Epoch: [252]  [ 10/312]  eta: 0:08:42  lr: 0.000284  min_lr: 0.000284  loss: 3.5072 (3.5612)  weight_decay: 0.0500 (0.0500)  time: 1.7315  data: 1.0404  max mem: 21002
Epoch: [252]  [ 20/312]  eta: 0:05:23  lr: 0.000283  min_lr: 0.000283  loss: 3.7896 (3.6447)  weight_decay: 0.0500 (0.0500)  time: 0.5501  data: 0.1606  max mem: 21002
Epoch: [252]  [ 30/312]  eta: 0:03:57  lr: 0.000283  min_lr: 0.000283  loss: 3.8168 (3.5689)  weight_decay: 0.0500 (0.0500)  time: 0.3557  data: 0.0498  max mem: 21002
Epoch: [252]  [ 40/312]  eta: 0:03:13  lr: 0.000283  min_lr: 0.000283  loss: 3.1919 (3.5072)  weight_decay: 0.0500 (0.0500)  time: 0.2936  data: 0.0015  max mem: 21002
Epoch: [252]  [ 50/312]  eta: 0:03:03  lr: 0.000282  min_lr: 0.000282  loss: 3.7013 (3.5777)  weight_decay: 0.0500 (0.0500)  time: 0.4743  data: 0.0327  max mem: 21002
Epoch: [252]  [ 60/312]  eta: 0:02:43  lr: 0.000282  min_lr: 0.000282  loss: 3.8855 (3.5709)  weight_decay: 0.0500 (0.0500)  time: 0.5188  data: 0.0845  max mem: 21002
Epoch: [252]  [ 70/312]  eta: 0:02:37  lr: 0.000281  min_lr: 0.000281  loss: 3.7078 (3.5655)  weight_decay: 0.0500 (0.0500)  time: 0.5227  data: 0.1256  max mem: 21002
Epoch: [252]  [ 80/312]  eta: 0:02:29  lr: 0.000281  min_lr: 0.000281  loss: 3.5925 (3.5603)  weight_decay: 0.0500 (0.0500)  time: 0.6416  data: 0.1076  max mem: 21002
Epoch: [252]  [ 90/312]  eta: 0:02:18  lr: 0.000281  min_lr: 0.000281  loss: 3.6110 (3.5709)  weight_decay: 0.0500 (0.0500)  time: 0.5250  data: 0.0731  max mem: 21002
Epoch: [252]  [100/312]  eta: 0:02:11  lr: 0.000280  min_lr: 0.000280  loss: 3.7106 (3.5826)  weight_decay: 0.0500 (0.0500)  time: 0.5281  data: 0.0614  max mem: 21002
Epoch: [252]  [110/312]  eta: 0:02:00  lr: 0.000280  min_lr: 0.000280  loss: 3.6446 (3.5736)  weight_decay: 0.0500 (0.0500)  time: 0.4710  data: 0.0235  max mem: 21002
Epoch: [252]  [120/312]  eta: 0:01:54  lr: 0.000280  min_lr: 0.000280  loss: 3.6050 (3.5569)  weight_decay: 0.0500 (0.0500)  time: 0.4747  data: 0.0397  max mem: 21002
Epoch: [252]  [130/312]  eta: 0:01:49  lr: 0.000279  min_lr: 0.000279  loss: 3.5874 (3.5556)  weight_decay: 0.0500 (0.0500)  time: 0.6397  data: 0.0729  max mem: 21002
Epoch: [252]  [140/312]  eta: 0:01:40  lr: 0.000279  min_lr: 0.000279  loss: 3.6923 (3.5706)  weight_decay: 0.0500 (0.0500)  time: 0.5118  data: 0.0696  max mem: 21002
Epoch: [252]  [150/312]  eta: 0:01:36  lr: 0.000279  min_lr: 0.000279  loss: 3.6923 (3.5706)  weight_decay: 0.0500 (0.0500)  time: 0.5383  data: 0.0749  max mem: 21002
Epoch: [252]  [160/312]  eta: 0:01:30  lr: 0.000278  min_lr: 0.000278  loss: 3.6150 (3.5790)  weight_decay: 0.0500 (0.0500)  time: 0.6407  data: 0.0678  max mem: 21002
Epoch: [252]  [170/312]  eta: 0:01:22  lr: 0.000278  min_lr: 0.000278  loss: 3.6086 (3.5685)  weight_decay: 0.0500 (0.0500)  time: 0.4697  data: 0.0487  max mem: 21002
Epoch: [252]  [180/312]  eta: 0:01:17  lr: 0.000277  min_lr: 0.000277  loss: 3.4173 (3.5622)  weight_decay: 0.0500 (0.0500)  time: 0.5373  data: 0.0738  max mem: 21002
Epoch: [252]  [190/312]  eta: 0:01:09  lr: 0.000277  min_lr: 0.000277  loss: 3.7781 (3.5683)  weight_decay: 0.0500 (0.0500)  time: 0.5098  data: 0.0596  max mem: 21002
Epoch: [252]  [200/312]  eta: 0:01:05  lr: 0.000277  min_lr: 0.000277  loss: 3.8230 (3.5689)  weight_decay: 0.0500 (0.0500)  time: 0.5400  data: 0.0512  max mem: 21002
Epoch: [252]  [210/312]  eta: 0:00:59  lr: 0.000276  min_lr: 0.000276  loss: 3.4201 (3.5629)  weight_decay: 0.0500 (0.0500)  time: 0.6904  data: 0.0453  max mem: 21002
Epoch: [252]  [220/312]  eta: 0:00:52  lr: 0.000276  min_lr: 0.000276  loss: 3.3248 (3.5499)  weight_decay: 0.0500 (0.0500)  time: 0.4785  data: 0.0350  max mem: 21002
Epoch: [252]  [230/312]  eta: 0:00:47  lr: 0.000276  min_lr: 0.000276  loss: 3.3713 (3.5498)  weight_decay: 0.0500 (0.0500)  time: 0.5694  data: 0.0804  max mem: 21002
Epoch: [252]  [240/312]  eta: 0:00:41  lr: 0.000275  min_lr: 0.000275  loss: 3.6038 (3.5523)  weight_decay: 0.0500 (0.0500)  time: 0.6252  data: 0.0461  max mem: 21002
Epoch: [252]  [250/312]  eta: 0:00:35  lr: 0.000275  min_lr: 0.000275  loss: 3.8363 (3.5624)  weight_decay: 0.0500 (0.0500)  time: 0.4487  data: 0.0489  max mem: 21002
Epoch: [252]  [260/312]  eta: 0:00:29  lr: 0.000275  min_lr: 0.000275  loss: 3.8363 (3.5658)  weight_decay: 0.0500 (0.0500)  time: 0.5674  data: 0.0863  max mem: 21002
Epoch: [252]  [270/312]  eta: 0:00:23  lr: 0.000274  min_lr: 0.000274  loss: 3.7453 (3.5647)  weight_decay: 0.0500 (0.0500)  time: 0.4968  data: 0.0382  max mem: 21002
Epoch: [252]  [280/312]  eta: 0:00:18  lr: 0.000274  min_lr: 0.000274  loss: 3.5112 (3.5645)  weight_decay: 0.0500 (0.0500)  time: 0.5370  data: 0.0440  max mem: 21002
Epoch: [252]  [290/312]  eta: 0:00:12  lr: 0.000273  min_lr: 0.000273  loss: 3.9603 (3.5783)  weight_decay: 0.0500 (0.0500)  time: 0.6278  data: 0.0437  max mem: 21002
Epoch: [252]  [300/312]  eta: 0:00:06  lr: 0.000273  min_lr: 0.000273  loss: 3.9356 (3.5830)  weight_decay: 0.0500 (0.0500)  time: 0.3842  data: 0.0114  max mem: 21002
Epoch: [252]  [310/312]  eta: 0:00:01  lr: 0.000273  min_lr: 0.000273  loss: 3.8843 (3.5884)  weight_decay: 0.0500 (0.0500)  time: 0.2875  data: 0.0113  max mem: 21002
Epoch: [252]  [311/312]  eta: 0:00:00  lr: 0.000273  min_lr: 0.000273  loss: 3.8843 (3.5909)  weight_decay: 0.0500 (0.0500)  time: 0.2768  data: 0.0001  max mem: 21002
Epoch: [252] Total time: 0:02:53 (0.5546 s / it)
Averaged stats: lr: 0.000273  min_lr: 0.000273  loss: 3.8843 (3.6302)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 1.1315 (1.1315)  acc1: 79.9479 (79.9479)  acc5: 95.1823 (95.1823)  time: 8.1459  data: 8.0290  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.5054 (1.4405)  acc1: 70.9635 (71.8720)  acc5: 90.3646 (90.6880)  time: 1.0032  data: 0.9032  max mem: 21002
Test: Total time: 0:00:09 (1.0142 s / it)
* Acc@1 71.392 Acc@5 90.546 loss 1.453
Accuracy of the model on the 50000 test images: 71.4%
Max accuracy: 72.08%
Epoch: [253]  [  0/312]  eta: 1:04:46  lr: 0.000273  min_lr: 0.000273  loss: 3.9073 (3.9073)  weight_decay: 0.0500 (0.0500)  time: 12.4572  data: 12.1594  max mem: 21002
Epoch: [253]  [ 10/312]  eta: 0:08:24  lr: 0.000272  min_lr: 0.000272  loss: 3.9073 (3.6394)  weight_decay: 0.0500 (0.0500)  time: 1.6702  data: 1.1061  max mem: 21002
Epoch: [253]  [ 20/312]  eta: 0:05:09  lr: 0.000272  min_lr: 0.000272  loss: 3.7884 (3.6085)  weight_decay: 0.0500 (0.0500)  time: 0.4900  data: 0.0464  max mem: 21002
Epoch: [253]  [ 30/312]  eta: 0:03:49  lr: 0.000272  min_lr: 0.000272  loss: 3.7714 (3.6532)  weight_decay: 0.0500 (0.0500)  time: 0.3408  data: 0.0463  max mem: 21002
Epoch: [253]  [ 40/312]  eta: 0:03:10  lr: 0.000271  min_lr: 0.000271  loss: 3.7714 (3.6316)  weight_decay: 0.0500 (0.0500)  time: 0.3219  data: 0.0246  max mem: 21002
Epoch: [253]  [ 50/312]  eta: 0:03:03  lr: 0.000271  min_lr: 0.000271  loss: 3.7593 (3.6269)  weight_decay: 0.0500 (0.0500)  time: 0.5271  data: 0.2004  max mem: 21002
Epoch: [253]  [ 60/312]  eta: 0:02:39  lr: 0.000270  min_lr: 0.000270  loss: 3.7593 (3.6397)  weight_decay: 0.0500 (0.0500)  time: 0.4956  data: 0.1765  max mem: 21002
Epoch: [253]  [ 70/312]  eta: 0:02:35  lr: 0.000270  min_lr: 0.000270  loss: 3.7984 (3.6608)  weight_decay: 0.0500 (0.0500)  time: 0.4978  data: 0.1529  max mem: 21002
Epoch: [253]  [ 80/312]  eta: 0:02:30  lr: 0.000270  min_lr: 0.000270  loss: 3.6572 (3.6238)  weight_decay: 0.0500 (0.0500)  time: 0.6901  data: 0.2855  max mem: 21002
Epoch: [253]  [ 90/312]  eta: 0:02:14  lr: 0.000269  min_lr: 0.000269  loss: 3.4183 (3.5976)  weight_decay: 0.0500 (0.0500)  time: 0.4822  data: 0.1338  max mem: 21002
Epoch: [253]  [100/312]  eta: 0:02:11  lr: 0.000269  min_lr: 0.000269  loss: 3.0120 (3.5459)  weight_decay: 0.0500 (0.0500)  time: 0.5020  data: 0.1379  max mem: 21002
Epoch: [253]  [110/312]  eta: 0:01:58  lr: 0.000269  min_lr: 0.000269  loss: 3.2629 (3.5627)  weight_decay: 0.0500 (0.0500)  time: 0.4984  data: 0.1375  max mem: 21002
Epoch: [253]  [120/312]  eta: 0:01:54  lr: 0.000268  min_lr: 0.000268  loss: 3.8386 (3.5692)  weight_decay: 0.0500 (0.0500)  time: 0.4847  data: 0.1284  max mem: 21002
Epoch: [253]  [130/312]  eta: 0:01:50  lr: 0.000268  min_lr: 0.000268  loss: 3.6875 (3.5792)  weight_decay: 0.0500 (0.0500)  time: 0.7192  data: 0.2326  max mem: 21002
Epoch: [253]  [140/312]  eta: 0:01:40  lr: 0.000268  min_lr: 0.000268  loss: 3.7897 (3.5933)  weight_decay: 0.0500 (0.0500)  time: 0.5212  data: 0.1048  max mem: 21002
Epoch: [253]  [150/312]  eta: 0:01:36  lr: 0.000267  min_lr: 0.000267  loss: 3.8275 (3.6084)  weight_decay: 0.0500 (0.0500)  time: 0.5331  data: 0.0854  max mem: 21002
Epoch: [253]  [160/312]  eta: 0:01:31  lr: 0.000267  min_lr: 0.000267  loss: 3.7513 (3.6102)  weight_decay: 0.0500 (0.0500)  time: 0.7370  data: 0.1266  max mem: 21002
Epoch: [253]  [170/312]  eta: 0:01:23  lr: 0.000267  min_lr: 0.000267  loss: 3.5261 (3.5948)  weight_decay: 0.0500 (0.0500)  time: 0.4974  data: 0.0420  max mem: 21002
Epoch: [253]  [180/312]  eta: 0:01:19  lr: 0.000266  min_lr: 0.000266  loss: 3.4621 (3.6022)  weight_decay: 0.0500 (0.0500)  time: 0.5539  data: 0.0109  max mem: 21002
Epoch: [253]  [190/312]  eta: 0:01:11  lr: 0.000266  min_lr: 0.000266  loss: 3.5588 (3.5999)  weight_decay: 0.0500 (0.0500)  time: 0.5633  data: 0.0109  max mem: 21002
Epoch: [253]  [200/312]  eta: 0:01:05  lr: 0.000265  min_lr: 0.000265  loss: 3.5340 (3.5942)  weight_decay: 0.0500 (0.0500)  time: 0.4431  data: 0.0461  max mem: 21002
Epoch: [253]  [210/312]  eta: 0:00:59  lr: 0.000265  min_lr: 0.000265  loss: 3.7271 (3.5925)  weight_decay: 0.0500 (0.0500)  time: 0.6252  data: 0.0698  max mem: 21002
Epoch: [253]  [220/312]  eta: 0:00:53  lr: 0.000265  min_lr: 0.000265  loss: 3.7271 (3.5904)  weight_decay: 0.0500 (0.0500)  time: 0.5466  data: 0.0246  max mem: 21002
Epoch: [253]  [230/312]  eta: 0:00:47  lr: 0.000264  min_lr: 0.000264  loss: 3.5916 (3.5885)  weight_decay: 0.0500 (0.0500)  time: 0.4948  data: 0.0324  max mem: 21002
Epoch: [253]  [240/312]  eta: 0:00:42  lr: 0.000264  min_lr: 0.000264  loss: 3.5916 (3.5871)  weight_decay: 0.0500 (0.0500)  time: 0.6319  data: 0.0792  max mem: 21002
Epoch: [253]  [250/312]  eta: 0:00:35  lr: 0.000264  min_lr: 0.000264  loss: 3.6617 (3.5771)  weight_decay: 0.0500 (0.0500)  time: 0.5224  data: 0.0476  max mem: 21002
Epoch: [253]  [260/312]  eta: 0:00:30  lr: 0.000263  min_lr: 0.000263  loss: 3.6705 (3.5826)  weight_decay: 0.0500 (0.0500)  time: 0.5395  data: 0.0009  max mem: 21002
Epoch: [253]  [270/312]  eta: 0:00:23  lr: 0.000263  min_lr: 0.000263  loss: 3.8532 (3.5850)  weight_decay: 0.0500 (0.0500)  time: 0.5045  data: 0.0009  max mem: 21002
Epoch: [253]  [280/312]  eta: 0:00:18  lr: 0.000263  min_lr: 0.000263  loss: 3.4942 (3.5764)  weight_decay: 0.0500 (0.0500)  time: 0.4597  data: 0.0621  max mem: 21002
Epoch: [253]  [290/312]  eta: 0:00:12  lr: 0.000262  min_lr: 0.000262  loss: 3.1071 (3.5723)  weight_decay: 0.0500 (0.0500)  time: 0.5496  data: 0.0849  max mem: 21002
Epoch: [253]  [300/312]  eta: 0:00:06  lr: 0.000262  min_lr: 0.000262  loss: 3.4824 (3.5686)  weight_decay: 0.0500 (0.0500)  time: 0.3809  data: 0.0232  max mem: 21002
Epoch: [253]  [310/312]  eta: 0:00:01  lr: 0.000262  min_lr: 0.000262  loss: 3.7386 (3.5759)  weight_decay: 0.0500 (0.0500)  time: 0.2851  data: 0.0001  max mem: 21002
Epoch: [253]  [311/312]  eta: 0:00:00  lr: 0.000261  min_lr: 0.000261  loss: 3.7386 (3.5774)  weight_decay: 0.0500 (0.0500)  time: 0.2768  data: 0.0001  max mem: 21002
Epoch: [253] Total time: 0:02:52 (0.5534 s / it)
Averaged stats: lr: 0.000261  min_lr: 0.000261  loss: 3.7386 (3.6363)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.1525 (1.1525)  acc1: 78.9062 (78.9062)  acc5: 94.7917 (94.7917)  time: 8.3537  data: 8.2349  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.5081 (1.4207)  acc1: 71.7448 (72.5920)  acc5: 90.6250 (90.9920)  time: 1.0228  data: 0.9151  max mem: 21002
Test: Total time: 0:00:09 (1.0508 s / it)
* Acc@1 72.088 Acc@5 91.132 loss 1.426
Accuracy of the model on the 50000 test images: 72.1%
Max accuracy: 72.09%
Epoch: [254]  [  0/312]  eta: 0:55:55  lr: 0.000261  min_lr: 0.000261  loss: 2.7564 (2.7564)  weight_decay: 0.0500 (0.0500)  time: 10.7546  data: 9.9672  max mem: 21002
Epoch: [254]  [ 10/312]  eta: 0:07:19  lr: 0.000261  min_lr: 0.000261  loss: 3.0281 (3.3236)  weight_decay: 0.0500 (0.0500)  time: 1.4564  data: 1.0872  max mem: 21002
Epoch: [254]  [ 20/312]  eta: 0:05:05  lr: 0.000261  min_lr: 0.000261  loss: 3.1639 (3.3721)  weight_decay: 0.0500 (0.0500)  time: 0.5612  data: 0.2403  max mem: 21002
Epoch: [254]  [ 30/312]  eta: 0:03:46  lr: 0.000260  min_lr: 0.000260  loss: 3.3154 (3.3580)  weight_decay: 0.0500 (0.0500)  time: 0.4412  data: 0.1411  max mem: 21002
Epoch: [254]  [ 40/312]  eta: 0:03:37  lr: 0.000260  min_lr: 0.000260  loss: 3.6715 (3.4714)  weight_decay: 0.0500 (0.0500)  time: 0.5432  data: 0.1633  max mem: 21002
Epoch: [254]  [ 50/312]  eta: 0:03:16  lr: 0.000260  min_lr: 0.000260  loss: 3.7255 (3.5122)  weight_decay: 0.0500 (0.0500)  time: 0.6716  data: 0.2865  max mem: 21002
Epoch: [254]  [ 60/312]  eta: 0:02:49  lr: 0.000259  min_lr: 0.000259  loss: 3.7255 (3.5274)  weight_decay: 0.0500 (0.0500)  time: 0.4139  data: 0.1238  max mem: 21002
Epoch: [254]  [ 70/312]  eta: 0:02:41  lr: 0.000259  min_lr: 0.000259  loss: 3.5399 (3.4759)  weight_decay: 0.0500 (0.0500)  time: 0.4487  data: 0.1549  max mem: 21002
Epoch: [254]  [ 80/312]  eta: 0:02:31  lr: 0.000259  min_lr: 0.000259  loss: 3.0664 (3.4668)  weight_decay: 0.0500 (0.0500)  time: 0.5957  data: 0.3004  max mem: 21002
Epoch: [254]  [ 90/312]  eta: 0:02:18  lr: 0.000258  min_lr: 0.000258  loss: 3.3518 (3.4498)  weight_decay: 0.0500 (0.0500)  time: 0.4742  data: 0.1895  max mem: 21002
Epoch: [254]  [100/312]  eta: 0:02:12  lr: 0.000258  min_lr: 0.000258  loss: 3.7722 (3.4786)  weight_decay: 0.0500 (0.0500)  time: 0.5055  data: 0.2080  max mem: 21002
Epoch: [254]  [110/312]  eta: 0:02:00  lr: 0.000258  min_lr: 0.000258  loss: 3.8747 (3.5019)  weight_decay: 0.0500 (0.0500)  time: 0.4649  data: 0.1645  max mem: 21002
Epoch: [254]  [120/312]  eta: 0:01:55  lr: 0.000257  min_lr: 0.000257  loss: 3.8747 (3.5180)  weight_decay: 0.0500 (0.0500)  time: 0.4704  data: 0.1836  max mem: 21002
Epoch: [254]  [130/312]  eta: 0:01:49  lr: 0.000257  min_lr: 0.000257  loss: 3.6386 (3.5167)  weight_decay: 0.0500 (0.0500)  time: 0.6435  data: 0.3541  max mem: 21002
Epoch: [254]  [140/312]  eta: 0:01:39  lr: 0.000257  min_lr: 0.000257  loss: 3.3431 (3.5187)  weight_decay: 0.0500 (0.0500)  time: 0.4658  data: 0.1711  max mem: 21002
Epoch: [254]  [150/312]  eta: 0:01:35  lr: 0.000256  min_lr: 0.000256  loss: 3.6202 (3.5382)  weight_decay: 0.0500 (0.0500)  time: 0.5041  data: 0.1969  max mem: 21002
Epoch: [254]  [160/312]  eta: 0:01:29  lr: 0.000256  min_lr: 0.000256  loss: 3.6818 (3.5419)  weight_decay: 0.0500 (0.0500)  time: 0.6312  data: 0.3022  max mem: 21002
Epoch: [254]  [170/312]  eta: 0:01:22  lr: 0.000255  min_lr: 0.000255  loss: 3.7814 (3.5536)  weight_decay: 0.0500 (0.0500)  time: 0.4921  data: 0.1651  max mem: 21002
Epoch: [254]  [180/312]  eta: 0:01:17  lr: 0.000255  min_lr: 0.000255  loss: 3.7814 (3.5541)  weight_decay: 0.0500 (0.0500)  time: 0.5699  data: 0.2286  max mem: 21002
Epoch: [254]  [190/312]  eta: 0:01:09  lr: 0.000255  min_lr: 0.000255  loss: 3.6043 (3.5664)  weight_decay: 0.0500 (0.0500)  time: 0.4987  data: 0.1696  max mem: 21002
Epoch: [254]  [200/312]  eta: 0:01:04  lr: 0.000254  min_lr: 0.000254  loss: 3.6713 (3.5701)  weight_decay: 0.0500 (0.0500)  time: 0.4866  data: 0.1473  max mem: 21002
Epoch: [254]  [210/312]  eta: 0:00:58  lr: 0.000254  min_lr: 0.000254  loss: 3.6799 (3.5729)  weight_decay: 0.0500 (0.0500)  time: 0.6294  data: 0.2700  max mem: 21002
Epoch: [254]  [220/312]  eta: 0:00:52  lr: 0.000254  min_lr: 0.000254  loss: 3.6462 (3.5689)  weight_decay: 0.0500 (0.0500)  time: 0.4882  data: 0.1235  max mem: 21002
Epoch: [254]  [230/312]  eta: 0:00:46  lr: 0.000253  min_lr: 0.000253  loss: 3.7140 (3.5781)  weight_decay: 0.0500 (0.0500)  time: 0.5402  data: 0.1026  max mem: 21002
Epoch: [254]  [240/312]  eta: 0:00:41  lr: 0.000253  min_lr: 0.000253  loss: 3.8858 (3.5847)  weight_decay: 0.0500 (0.0500)  time: 0.6078  data: 0.1587  max mem: 21002
Epoch: [254]  [250/312]  eta: 0:00:35  lr: 0.000253  min_lr: 0.000253  loss: 3.8189 (3.5824)  weight_decay: 0.0500 (0.0500)  time: 0.5193  data: 0.1128  max mem: 21002
Epoch: [254]  [260/312]  eta: 0:00:29  lr: 0.000252  min_lr: 0.000252  loss: 3.9428 (3.5888)  weight_decay: 0.0500 (0.0500)  time: 0.6034  data: 0.0910  max mem: 21002
Epoch: [254]  [270/312]  eta: 0:00:23  lr: 0.000252  min_lr: 0.000252  loss: 3.5999 (3.5789)  weight_decay: 0.0500 (0.0500)  time: 0.4961  data: 0.0349  max mem: 21002
Epoch: [254]  [280/312]  eta: 0:00:18  lr: 0.000252  min_lr: 0.000252  loss: 3.7747 (3.5853)  weight_decay: 0.0500 (0.0500)  time: 0.4811  data: 0.0414  max mem: 21002
Epoch: [254]  [290/312]  eta: 0:00:12  lr: 0.000251  min_lr: 0.000251  loss: 3.9252 (3.5971)  weight_decay: 0.0500 (0.0500)  time: 0.5363  data: 0.0518  max mem: 21002
Epoch: [254]  [300/312]  eta: 0:00:06  lr: 0.000251  min_lr: 0.000251  loss: 3.9325 (3.6046)  weight_decay: 0.0500 (0.0500)  time: 0.3653  data: 0.0108  max mem: 21002
Epoch: [254]  [310/312]  eta: 0:00:01  lr: 0.000251  min_lr: 0.000251  loss: 3.8661 (3.5991)  weight_decay: 0.0500 (0.0500)  time: 0.3041  data: 0.0001  max mem: 21002
Epoch: [254]  [311/312]  eta: 0:00:00  lr: 0.000251  min_lr: 0.000251  loss: 3.5894 (3.5975)  weight_decay: 0.0500 (0.0500)  time: 0.2771  data: 0.0001  max mem: 21002
Epoch: [254] Total time: 0:02:50 (0.5472 s / it)
Averaged stats: lr: 0.000251  min_lr: 0.000251  loss: 3.5894 (3.6142)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.0520 (1.0520)  acc1: 80.2083 (80.2083)  acc5: 95.5729 (95.5729)  time: 8.5384  data: 8.4201  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4028 (1.3186)  acc1: 72.0052 (72.5600)  acc5: 90.8854 (91.3440)  time: 1.0376  data: 0.9356  max mem: 21002
Test: Total time: 0:00:09 (1.0538 s / it)
* Acc@1 72.316 Acc@5 91.334 loss 1.327
Accuracy of the model on the 50000 test images: 72.3%
Max accuracy: 72.32%
Epoch: [255]  [  0/312]  eta: 0:57:44  lr: 0.000250  min_lr: 0.000250  loss: 4.1130 (4.1130)  weight_decay: 0.0500 (0.0500)  time: 11.1056  data: 8.4944  max mem: 21002
Epoch: [255]  [ 10/312]  eta: 0:07:57  lr: 0.000250  min_lr: 0.000250  loss: 3.7792 (3.5560)  weight_decay: 0.0500 (0.0500)  time: 1.5804  data: 1.0856  max mem: 21002
Epoch: [255]  [ 20/312]  eta: 0:05:23  lr: 0.000250  min_lr: 0.000250  loss: 3.3233 (3.5369)  weight_decay: 0.0500 (0.0500)  time: 0.6085  data: 0.2479  max mem: 21002
Epoch: [255]  [ 30/312]  eta: 0:03:58  lr: 0.000249  min_lr: 0.000249  loss: 3.6070 (3.6627)  weight_decay: 0.0500 (0.0500)  time: 0.4390  data: 0.0758  max mem: 21002
Epoch: [255]  [ 40/312]  eta: 0:03:17  lr: 0.000249  min_lr: 0.000249  loss: 3.8364 (3.6651)  weight_decay: 0.0500 (0.0500)  time: 0.3214  data: 0.0243  max mem: 21002
Epoch: [255]  [ 50/312]  eta: 0:03:06  lr: 0.000249  min_lr: 0.000249  loss: 3.7578 (3.6394)  weight_decay: 0.0500 (0.0500)  time: 0.5058  data: 0.1776  max mem: 21002
Epoch: [255]  [ 60/312]  eta: 0:02:42  lr: 0.000248  min_lr: 0.000248  loss: 3.7724 (3.6731)  weight_decay: 0.0500 (0.0500)  time: 0.4824  data: 0.1659  max mem: 21002
Epoch: [255]  [ 70/312]  eta: 0:02:35  lr: 0.000248  min_lr: 0.000248  loss: 3.6561 (3.6180)  weight_decay: 0.0500 (0.0500)  time: 0.4702  data: 0.1576  max mem: 21002
Epoch: [255]  [ 80/312]  eta: 0:02:28  lr: 0.000248  min_lr: 0.000248  loss: 3.6022 (3.6420)  weight_decay: 0.0500 (0.0500)  time: 0.6271  data: 0.2698  max mem: 21002
Epoch: [255]  [ 90/312]  eta: 0:02:16  lr: 0.000247  min_lr: 0.000247  loss: 3.8158 (3.6563)  weight_decay: 0.0500 (0.0500)  time: 0.5105  data: 0.1859  max mem: 21002
Epoch: [255]  [100/312]  eta: 0:02:12  lr: 0.000247  min_lr: 0.000247  loss: 3.6687 (3.6503)  weight_decay: 0.0500 (0.0500)  time: 0.5583  data: 0.2178  max mem: 21002
Epoch: [255]  [110/312]  eta: 0:01:59  lr: 0.000247  min_lr: 0.000247  loss: 3.6348 (3.6361)  weight_decay: 0.0500 (0.0500)  time: 0.5013  data: 0.1566  max mem: 21002
Epoch: [255]  [120/312]  eta: 0:01:55  lr: 0.000246  min_lr: 0.000246  loss: 3.5580 (3.6328)  weight_decay: 0.0500 (0.0500)  time: 0.4975  data: 0.1613  max mem: 21002
Epoch: [255]  [130/312]  eta: 0:01:49  lr: 0.000246  min_lr: 0.000246  loss: 3.5055 (3.6222)  weight_decay: 0.0500 (0.0500)  time: 0.6594  data: 0.2850  max mem: 21002
Epoch: [255]  [140/312]  eta: 0:01:40  lr: 0.000246  min_lr: 0.000246  loss: 3.6629 (3.6351)  weight_decay: 0.0500 (0.0500)  time: 0.4600  data: 0.1373  max mem: 21002
Epoch: [255]  [150/312]  eta: 0:01:36  lr: 0.000245  min_lr: 0.000245  loss: 3.7912 (3.6398)  weight_decay: 0.0500 (0.0500)  time: 0.5378  data: 0.1631  max mem: 21002
Epoch: [255]  [160/312]  eta: 0:01:29  lr: 0.000245  min_lr: 0.000245  loss: 3.7318 (3.6363)  weight_decay: 0.0500 (0.0500)  time: 0.6430  data: 0.1701  max mem: 21002
Epoch: [255]  [170/312]  eta: 0:01:23  lr: 0.000245  min_lr: 0.000245  loss: 3.9041 (3.6520)  weight_decay: 0.0500 (0.0500)  time: 0.5072  data: 0.1240  max mem: 21002
Epoch: [255]  [180/312]  eta: 0:01:17  lr: 0.000244  min_lr: 0.000244  loss: 3.9140 (3.6506)  weight_decay: 0.0500 (0.0500)  time: 0.5681  data: 0.1992  max mem: 21002
Epoch: [255]  [190/312]  eta: 0:01:09  lr: 0.000244  min_lr: 0.000244  loss: 3.4920 (3.6360)  weight_decay: 0.0500 (0.0500)  time: 0.4669  data: 0.0960  max mem: 21002
Epoch: [255]  [200/312]  eta: 0:01:04  lr: 0.000244  min_lr: 0.000244  loss: 3.2415 (3.6235)  weight_decay: 0.0500 (0.0500)  time: 0.4800  data: 0.1296  max mem: 21002
Epoch: [255]  [210/312]  eta: 0:00:59  lr: 0.000243  min_lr: 0.000243  loss: 3.6743 (3.6228)  weight_decay: 0.0500 (0.0500)  time: 0.6506  data: 0.2611  max mem: 21002
Epoch: [255]  [220/312]  eta: 0:00:52  lr: 0.000243  min_lr: 0.000243  loss: 3.7866 (3.6236)  weight_decay: 0.0500 (0.0500)  time: 0.4599  data: 0.1322  max mem: 21002
Epoch: [255]  [230/312]  eta: 0:00:46  lr: 0.000243  min_lr: 0.000243  loss: 3.8323 (3.6354)  weight_decay: 0.0500 (0.0500)  time: 0.4578  data: 0.1629  max mem: 21002
Epoch: [255]  [240/312]  eta: 0:00:40  lr: 0.000242  min_lr: 0.000242  loss: 3.7954 (3.6275)  weight_decay: 0.0500 (0.0500)  time: 0.5865  data: 0.2903  max mem: 21002
Epoch: [255]  [250/312]  eta: 0:00:34  lr: 0.000242  min_lr: 0.000242  loss: 3.3393 (3.6243)  weight_decay: 0.0500 (0.0500)  time: 0.4751  data: 0.1890  max mem: 21002
Epoch: [255]  [260/312]  eta: 0:00:29  lr: 0.000242  min_lr: 0.000242  loss: 3.3073 (3.6214)  weight_decay: 0.0500 (0.0500)  time: 0.5428  data: 0.2589  max mem: 21002
Epoch: [255]  [270/312]  eta: 0:00:23  lr: 0.000241  min_lr: 0.000241  loss: 3.6740 (3.6251)  weight_decay: 0.0500 (0.0500)  time: 0.4832  data: 0.1979  max mem: 21002
Epoch: [255]  [280/312]  eta: 0:00:17  lr: 0.000241  min_lr: 0.000241  loss: 3.6740 (3.6151)  weight_decay: 0.0500 (0.0500)  time: 0.4965  data: 0.2053  max mem: 21002
Epoch: [255]  [290/312]  eta: 0:00:12  lr: 0.000240  min_lr: 0.000240  loss: 3.3269 (3.6116)  weight_decay: 0.0500 (0.0500)  time: 0.6219  data: 0.3315  max mem: 21002
Epoch: [255]  [300/312]  eta: 0:00:06  lr: 0.000240  min_lr: 0.000240  loss: 3.3594 (3.6088)  weight_decay: 0.0500 (0.0500)  time: 0.4155  data: 0.1350  max mem: 21002
Epoch: [255]  [310/312]  eta: 0:00:01  lr: 0.000240  min_lr: 0.000240  loss: 3.7579 (3.6141)  weight_decay: 0.0500 (0.0500)  time: 0.2921  data: 0.0161  max mem: 21002
Epoch: [255]  [311/312]  eta: 0:00:00  lr: 0.000240  min_lr: 0.000240  loss: 3.7579 (3.6161)  weight_decay: 0.0500 (0.0500)  time: 0.2841  data: 0.0075  max mem: 21002
Epoch: [255] Total time: 0:02:50 (0.5461 s / it)
Averaged stats: lr: 0.000240  min_lr: 0.000240  loss: 3.7579 (3.6253)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.1037 (1.1037)  acc1: 78.3854 (78.3854)  acc5: 95.0521 (95.0521)  time: 8.7106  data: 8.5935  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4469 (1.3403)  acc1: 71.2240 (72.0640)  acc5: 90.4948 (90.9920)  time: 1.0556  data: 0.9549  max mem: 21002
Test: Total time: 0:00:09 (1.0751 s / it)
* Acc@1 72.160 Acc@5 91.118 loss 1.340
Accuracy of the model on the 50000 test images: 72.2%
Max accuracy: 72.32%
Epoch: [256]  [  0/312]  eta: 0:59:44  lr: 0.000240  min_lr: 0.000240  loss: 2.7938 (2.7938)  weight_decay: 0.0500 (0.0500)  time: 11.4874  data: 11.0109  max mem: 21002
Epoch: [256]  [ 10/312]  eta: 0:07:25  lr: 0.000239  min_lr: 0.000239  loss: 3.3787 (3.3607)  weight_decay: 0.0500 (0.0500)  time: 1.4746  data: 1.0015  max mem: 21002
Epoch: [256]  [ 20/312]  eta: 0:05:00  lr: 0.000239  min_lr: 0.000239  loss: 3.8023 (3.5085)  weight_decay: 0.0500 (0.0500)  time: 0.5052  data: 0.0909  max mem: 21002
Epoch: [256]  [ 30/312]  eta: 0:03:42  lr: 0.000239  min_lr: 0.000239  loss: 3.8418 (3.6059)  weight_decay: 0.0500 (0.0500)  time: 0.4112  data: 0.0909  max mem: 21002
Epoch: [256]  [ 40/312]  eta: 0:03:21  lr: 0.000238  min_lr: 0.000238  loss: 3.9487 (3.5840)  weight_decay: 0.0500 (0.0500)  time: 0.4389  data: 0.1405  max mem: 21002
Epoch: [256]  [ 50/312]  eta: 0:03:11  lr: 0.000238  min_lr: 0.000238  loss: 3.9149 (3.5887)  weight_decay: 0.0500 (0.0500)  time: 0.6409  data: 0.3426  max mem: 21002
Epoch: [256]  [ 60/312]  eta: 0:02:45  lr: 0.000238  min_lr: 0.000238  loss: 3.8449 (3.5834)  weight_decay: 0.0500 (0.0500)  time: 0.4884  data: 0.2027  max mem: 21002
Epoch: [256]  [ 70/312]  eta: 0:02:39  lr: 0.000237  min_lr: 0.000237  loss: 3.8320 (3.6031)  weight_decay: 0.0500 (0.0500)  time: 0.4712  data: 0.1743  max mem: 21002
Epoch: [256]  [ 80/312]  eta: 0:02:32  lr: 0.000237  min_lr: 0.000237  loss: 3.6945 (3.5946)  weight_decay: 0.0500 (0.0500)  time: 0.6537  data: 0.3576  max mem: 21002
Epoch: [256]  [ 90/312]  eta: 0:02:16  lr: 0.000237  min_lr: 0.000237  loss: 3.5775 (3.5881)  weight_decay: 0.0500 (0.0500)  time: 0.4697  data: 0.1838  max mem: 21002
Epoch: [256]  [100/312]  eta: 0:02:11  lr: 0.000236  min_lr: 0.000236  loss: 3.7409 (3.6075)  weight_decay: 0.0500 (0.0500)  time: 0.4786  data: 0.1935  max mem: 21002
Epoch: [256]  [110/312]  eta: 0:01:59  lr: 0.000236  min_lr: 0.000236  loss: 3.9068 (3.6391)  weight_decay: 0.0500 (0.0500)  time: 0.4776  data: 0.1935  max mem: 21002
Epoch: [256]  [120/312]  eta: 0:01:54  lr: 0.000236  min_lr: 0.000236  loss: 3.9068 (3.6604)  weight_decay: 0.0500 (0.0500)  time: 0.4691  data: 0.1849  max mem: 21002
Epoch: [256]  [130/312]  eta: 0:01:49  lr: 0.000235  min_lr: 0.000235  loss: 3.8997 (3.6800)  weight_decay: 0.0500 (0.0500)  time: 0.6612  data: 0.3710  max mem: 21002
Epoch: [256]  [140/312]  eta: 0:01:39  lr: 0.000235  min_lr: 0.000235  loss: 3.6637 (3.6576)  weight_decay: 0.0500 (0.0500)  time: 0.4767  data: 0.1868  max mem: 21002
Epoch: [256]  [150/312]  eta: 0:01:35  lr: 0.000235  min_lr: 0.000235  loss: 3.3847 (3.6428)  weight_decay: 0.0500 (0.0500)  time: 0.5021  data: 0.2156  max mem: 21002
Epoch: [256]  [160/312]  eta: 0:01:30  lr: 0.000234  min_lr: 0.000234  loss: 3.5079 (3.6453)  weight_decay: 0.0500 (0.0500)  time: 0.6937  data: 0.4046  max mem: 21002
Epoch: [256]  [170/312]  eta: 0:01:21  lr: 0.000234  min_lr: 0.000234  loss: 3.7932 (3.6502)  weight_decay: 0.0500 (0.0500)  time: 0.4770  data: 0.1897  max mem: 21002
Epoch: [256]  [180/312]  eta: 0:01:16  lr: 0.000234  min_lr: 0.000234  loss: 3.9328 (3.6680)  weight_decay: 0.0500 (0.0500)  time: 0.4809  data: 0.1945  max mem: 21002
Epoch: [256]  [190/312]  eta: 0:01:09  lr: 0.000233  min_lr: 0.000233  loss: 3.9780 (3.6725)  weight_decay: 0.0500 (0.0500)  time: 0.5078  data: 0.1944  max mem: 21002
Epoch: [256]  [200/312]  eta: 0:01:04  lr: 0.000233  min_lr: 0.000233  loss: 3.6225 (3.6704)  weight_decay: 0.0500 (0.0500)  time: 0.5325  data: 0.1769  max mem: 21002
Epoch: [256]  [210/312]  eta: 0:00:58  lr: 0.000233  min_lr: 0.000233  loss: 3.6405 (3.6770)  weight_decay: 0.0500 (0.0500)  time: 0.6448  data: 0.3122  max mem: 21002
Epoch: [256]  [220/312]  eta: 0:00:52  lr: 0.000232  min_lr: 0.000232  loss: 3.6176 (3.6699)  weight_decay: 0.0500 (0.0500)  time: 0.4926  data: 0.1365  max mem: 21002
Epoch: [256]  [230/312]  eta: 0:00:46  lr: 0.000232  min_lr: 0.000232  loss: 3.6861 (3.6731)  weight_decay: 0.0500 (0.0500)  time: 0.4761  data: 0.1226  max mem: 21002
Epoch: [256]  [240/312]  eta: 0:00:41  lr: 0.000232  min_lr: 0.000232  loss: 3.6933 (3.6747)  weight_decay: 0.0500 (0.0500)  time: 0.6070  data: 0.2349  max mem: 21002
Epoch: [256]  [250/312]  eta: 0:00:35  lr: 0.000231  min_lr: 0.000231  loss: 3.7214 (3.6776)  weight_decay: 0.0500 (0.0500)  time: 0.5857  data: 0.1136  max mem: 21002
Epoch: [256]  [260/312]  eta: 0:00:29  lr: 0.000231  min_lr: 0.000231  loss: 3.8732 (3.6699)  weight_decay: 0.0500 (0.0500)  time: 0.4998  data: 0.1095  max mem: 21002
Epoch: [256]  [270/312]  eta: 0:00:23  lr: 0.000231  min_lr: 0.000231  loss: 3.7722 (3.6736)  weight_decay: 0.0500 (0.0500)  time: 0.4648  data: 0.1101  max mem: 21002
Epoch: [256]  [280/312]  eta: 0:00:18  lr: 0.000230  min_lr: 0.000230  loss: 3.7722 (3.6723)  weight_decay: 0.0500 (0.0500)  time: 0.5585  data: 0.1081  max mem: 21002
Epoch: [256]  [290/312]  eta: 0:00:12  lr: 0.000230  min_lr: 0.000230  loss: 3.8105 (3.6773)  weight_decay: 0.0500 (0.0500)  time: 0.5497  data: 0.1556  max mem: 21002
Epoch: [256]  [300/312]  eta: 0:00:06  lr: 0.000230  min_lr: 0.000230  loss: 3.9001 (3.6813)  weight_decay: 0.0500 (0.0500)  time: 0.3489  data: 0.0488  max mem: 21002
Epoch: [256]  [310/312]  eta: 0:00:01  lr: 0.000229  min_lr: 0.000229  loss: 3.8574 (3.6850)  weight_decay: 0.0500 (0.0500)  time: 0.2867  data: 0.0001  max mem: 21002
Epoch: [256]  [311/312]  eta: 0:00:00  lr: 0.000229  min_lr: 0.000229  loss: 3.8574 (3.6878)  weight_decay: 0.0500 (0.0500)  time: 0.2865  data: 0.0001  max mem: 21002
Epoch: [256] Total time: 0:02:50 (0.5454 s / it)
Averaged stats: lr: 0.000229  min_lr: 0.000229  loss: 3.8574 (3.6373)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.2048 (1.2048)  acc1: 79.4271 (79.4271)  acc5: 94.7917 (94.7917)  time: 8.2237  data: 8.1052  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.5609 (1.4834)  acc1: 71.2240 (72.4160)  acc5: 91.1458 (91.3440)  time: 1.0378  data: 0.9415  max mem: 21002
Test: Total time: 0:00:09 (1.0480 s / it)
* Acc@1 72.126 Acc@5 91.242 loss 1.486
Accuracy of the model on the 50000 test images: 72.1%
Max accuracy: 72.32%
Epoch: [257]  [  0/312]  eta: 1:01:37  lr: 0.000229  min_lr: 0.000229  loss: 3.0032 (3.0032)  weight_decay: 0.0500 (0.0500)  time: 11.8525  data: 8.8095  max mem: 21002
Epoch: [257]  [ 10/312]  eta: 0:07:54  lr: 0.000229  min_lr: 0.000229  loss: 3.7659 (3.5205)  weight_decay: 0.0500 (0.0500)  time: 1.5715  data: 1.0286  max mem: 21002
Epoch: [257]  [ 20/312]  eta: 0:05:06  lr: 0.000229  min_lr: 0.000229  loss: 3.8384 (3.6081)  weight_decay: 0.0500 (0.0500)  time: 0.5095  data: 0.1977  max mem: 21002
Epoch: [257]  [ 30/312]  eta: 0:03:46  lr: 0.000228  min_lr: 0.000228  loss: 3.7616 (3.5608)  weight_decay: 0.0500 (0.0500)  time: 0.3818  data: 0.0728  max mem: 21002
Epoch: [257]  [ 40/312]  eta: 0:03:30  lr: 0.000228  min_lr: 0.000228  loss: 3.6099 (3.5894)  weight_decay: 0.0500 (0.0500)  time: 0.4866  data: 0.0818  max mem: 21002
Epoch: [257]  [ 50/312]  eta: 0:03:12  lr: 0.000228  min_lr: 0.000228  loss: 3.8562 (3.6760)  weight_decay: 0.0500 (0.0500)  time: 0.6257  data: 0.1270  max mem: 21002
Epoch: [257]  [ 60/312]  eta: 0:02:50  lr: 0.000227  min_lr: 0.000227  loss: 3.8704 (3.6733)  weight_decay: 0.0500 (0.0500)  time: 0.4719  data: 0.0929  max mem: 21002
Epoch: [257]  [ 70/312]  eta: 0:02:42  lr: 0.000227  min_lr: 0.000227  loss: 3.7175 (3.6643)  weight_decay: 0.0500 (0.0500)  time: 0.5057  data: 0.1453  max mem: 21002
Epoch: [257]  [ 80/312]  eta: 0:02:29  lr: 0.000227  min_lr: 0.000227  loss: 3.5414 (3.6397)  weight_decay: 0.0500 (0.0500)  time: 0.5547  data: 0.1480  max mem: 21002
Epoch: [257]  [ 90/312]  eta: 0:02:18  lr: 0.000226  min_lr: 0.000226  loss: 3.6150 (3.6426)  weight_decay: 0.0500 (0.0500)  time: 0.4658  data: 0.1336  max mem: 21002
Epoch: [257]  [100/312]  eta: 0:02:13  lr: 0.000226  min_lr: 0.000226  loss: 3.6150 (3.6221)  weight_decay: 0.0500 (0.0500)  time: 0.5734  data: 0.2227  max mem: 21002
Epoch: [257]  [110/312]  eta: 0:02:01  lr: 0.000226  min_lr: 0.000226  loss: 3.7323 (3.6321)  weight_decay: 0.0500 (0.0500)  time: 0.4903  data: 0.1397  max mem: 21002
Epoch: [257]  [120/312]  eta: 0:01:57  lr: 0.000225  min_lr: 0.000225  loss: 3.8631 (3.6275)  weight_decay: 0.0500 (0.0500)  time: 0.5038  data: 0.1164  max mem: 21002
Epoch: [257]  [130/312]  eta: 0:01:51  lr: 0.000225  min_lr: 0.000225  loss: 3.1553 (3.5926)  weight_decay: 0.0500 (0.0500)  time: 0.6663  data: 0.1929  max mem: 21002
Epoch: [257]  [140/312]  eta: 0:01:42  lr: 0.000225  min_lr: 0.000225  loss: 3.5306 (3.6059)  weight_decay: 0.0500 (0.0500)  time: 0.4908  data: 0.1163  max mem: 21002
Epoch: [257]  [150/312]  eta: 0:01:37  lr: 0.000224  min_lr: 0.000224  loss: 3.8407 (3.6107)  weight_decay: 0.0500 (0.0500)  time: 0.5260  data: 0.1421  max mem: 21002
Epoch: [257]  [160/312]  eta: 0:01:29  lr: 0.000224  min_lr: 0.000224  loss: 3.8571 (3.6113)  weight_decay: 0.0500 (0.0500)  time: 0.5712  data: 0.1031  max mem: 21002
Epoch: [257]  [170/312]  eta: 0:01:23  lr: 0.000224  min_lr: 0.000224  loss: 3.8364 (3.6232)  weight_decay: 0.0500 (0.0500)  time: 0.4755  data: 0.1046  max mem: 21002
Epoch: [257]  [180/312]  eta: 0:01:17  lr: 0.000223  min_lr: 0.000223  loss: 3.8030 (3.6305)  weight_decay: 0.0500 (0.0500)  time: 0.5781  data: 0.1948  max mem: 21002
Epoch: [257]  [190/312]  eta: 0:01:09  lr: 0.000223  min_lr: 0.000223  loss: 3.8544 (3.6393)  weight_decay: 0.0500 (0.0500)  time: 0.4764  data: 0.0908  max mem: 21002
Epoch: [257]  [200/312]  eta: 0:01:04  lr: 0.000223  min_lr: 0.000223  loss: 3.8544 (3.6358)  weight_decay: 0.0500 (0.0500)  time: 0.4844  data: 0.1055  max mem: 21002
Epoch: [257]  [210/312]  eta: 0:00:59  lr: 0.000222  min_lr: 0.000222  loss: 3.4894 (3.6257)  weight_decay: 0.0500 (0.0500)  time: 0.6550  data: 0.1794  max mem: 21002
Epoch: [257]  [220/312]  eta: 0:00:52  lr: 0.000222  min_lr: 0.000222  loss: 3.3443 (3.6134)  weight_decay: 0.0500 (0.0500)  time: 0.4705  data: 0.0864  max mem: 21002
Epoch: [257]  [230/312]  eta: 0:00:46  lr: 0.000222  min_lr: 0.000222  loss: 3.3443 (3.5966)  weight_decay: 0.0500 (0.0500)  time: 0.4856  data: 0.1053  max mem: 21002
Epoch: [257]  [240/312]  eta: 0:00:41  lr: 0.000221  min_lr: 0.000221  loss: 3.3639 (3.5939)  weight_decay: 0.0500 (0.0500)  time: 0.5867  data: 0.0994  max mem: 21002
Epoch: [257]  [250/312]  eta: 0:00:35  lr: 0.000221  min_lr: 0.000221  loss: 3.6406 (3.5896)  weight_decay: 0.0500 (0.0500)  time: 0.4994  data: 0.0874  max mem: 21002
Epoch: [257]  [260/312]  eta: 0:00:29  lr: 0.000221  min_lr: 0.000221  loss: 3.7689 (3.5916)  weight_decay: 0.0500 (0.0500)  time: 0.5672  data: 0.1498  max mem: 21002
Epoch: [257]  [270/312]  eta: 0:00:23  lr: 0.000220  min_lr: 0.000220  loss: 3.7720 (3.5937)  weight_decay: 0.0500 (0.0500)  time: 0.4690  data: 0.0691  max mem: 21002
Epoch: [257]  [280/312]  eta: 0:00:18  lr: 0.000220  min_lr: 0.000220  loss: 3.7720 (3.5976)  weight_decay: 0.0500 (0.0500)  time: 0.4974  data: 0.1056  max mem: 21002
Epoch: [257]  [290/312]  eta: 0:00:12  lr: 0.000220  min_lr: 0.000220  loss: 3.8308 (3.6016)  weight_decay: 0.0500 (0.0500)  time: 0.6021  data: 0.1469  max mem: 21002
Epoch: [257]  [300/312]  eta: 0:00:06  lr: 0.000219  min_lr: 0.000219  loss: 3.8925 (3.6044)  weight_decay: 0.0500 (0.0500)  time: 0.4119  data: 0.0655  max mem: 21002
Epoch: [257]  [310/312]  eta: 0:00:01  lr: 0.000219  min_lr: 0.000219  loss: 3.6853 (3.5955)  weight_decay: 0.0500 (0.0500)  time: 0.3004  data: 0.0239  max mem: 21002
Epoch: [257]  [311/312]  eta: 0:00:00  lr: 0.000219  min_lr: 0.000219  loss: 3.6853 (3.5954)  weight_decay: 0.0500 (0.0500)  time: 0.2771  data: 0.0001  max mem: 21002
Epoch: [257] Total time: 0:02:51 (0.5488 s / it)
Averaged stats: lr: 0.000219  min_lr: 0.000219  loss: 3.6853 (3.6051)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.1084 (1.1084)  acc1: 78.9062 (78.9062)  acc5: 95.0521 (95.0521)  time: 8.4633  data: 8.3443  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4214 (1.3278)  acc1: 72.9167 (72.9120)  acc5: 91.1458 (91.3760)  time: 1.0255  data: 0.9272  max mem: 21002
Test: Total time: 0:00:09 (1.0595 s / it)
* Acc@1 72.694 Acc@5 91.366 loss 1.336
Accuracy of the model on the 50000 test images: 72.7%
Max accuracy: 72.69%
Epoch: [258]  [  0/312]  eta: 0:58:54  lr: 0.000219  min_lr: 0.000219  loss: 4.2753 (4.2753)  weight_decay: 0.0500 (0.0500)  time: 11.3275  data: 9.7593  max mem: 21002
Epoch: [258]  [ 10/312]  eta: 0:07:31  lr: 0.000219  min_lr: 0.000219  loss: 3.9915 (3.9612)  weight_decay: 0.0500 (0.0500)  time: 1.4943  data: 1.0074  max mem: 21002
Epoch: [258]  [ 20/312]  eta: 0:05:27  lr: 0.000218  min_lr: 0.000218  loss: 3.7935 (3.7249)  weight_decay: 0.0500 (0.0500)  time: 0.6126  data: 0.1595  max mem: 21002
Epoch: [258]  [ 30/312]  eta: 0:04:01  lr: 0.000218  min_lr: 0.000218  loss: 3.8829 (3.8017)  weight_decay: 0.0500 (0.0500)  time: 0.5057  data: 0.0937  max mem: 21002
Epoch: [258]  [ 40/312]  eta: 0:03:26  lr: 0.000218  min_lr: 0.000218  loss: 3.9693 (3.7809)  weight_decay: 0.0500 (0.0500)  time: 0.3800  data: 0.0055  max mem: 21002
Epoch: [258]  [ 50/312]  eta: 0:03:13  lr: 0.000217  min_lr: 0.000217  loss: 3.5463 (3.6805)  weight_decay: 0.0500 (0.0500)  time: 0.5598  data: 0.1327  max mem: 21002
Epoch: [258]  [ 60/312]  eta: 0:02:47  lr: 0.000217  min_lr: 0.000217  loss: 3.5309 (3.6699)  weight_decay: 0.0500 (0.0500)  time: 0.4726  data: 0.1279  max mem: 21002
Epoch: [258]  [ 70/312]  eta: 0:02:41  lr: 0.000217  min_lr: 0.000217  loss: 3.6350 (3.6545)  weight_decay: 0.0500 (0.0500)  time: 0.4890  data: 0.1452  max mem: 21002
Epoch: [258]  [ 80/312]  eta: 0:02:34  lr: 0.000216  min_lr: 0.000216  loss: 3.8381 (3.6553)  weight_decay: 0.0500 (0.0500)  time: 0.6716  data: 0.2977  max mem: 21002
Epoch: [258]  [ 90/312]  eta: 0:02:18  lr: 0.000216  min_lr: 0.000216  loss: 3.7174 (3.6242)  weight_decay: 0.0500 (0.0500)  time: 0.4703  data: 0.1532  max mem: 21002
Epoch: [258]  [100/312]  eta: 0:02:13  lr: 0.000216  min_lr: 0.000216  loss: 3.1730 (3.5917)  weight_decay: 0.0500 (0.0500)  time: 0.4683  data: 0.0946  max mem: 21002
Epoch: [258]  [110/312]  eta: 0:02:00  lr: 0.000215  min_lr: 0.000215  loss: 3.3712 (3.5768)  weight_decay: 0.0500 (0.0500)  time: 0.4698  data: 0.0947  max mem: 21002
Epoch: [258]  [120/312]  eta: 0:01:56  lr: 0.000215  min_lr: 0.000215  loss: 3.6248 (3.5879)  weight_decay: 0.0500 (0.0500)  time: 0.4866  data: 0.1070  max mem: 21002
Epoch: [258]  [130/312]  eta: 0:01:51  lr: 0.000215  min_lr: 0.000215  loss: 3.8684 (3.5874)  weight_decay: 0.0500 (0.0500)  time: 0.6804  data: 0.2345  max mem: 21002
Epoch: [258]  [140/312]  eta: 0:01:41  lr: 0.000214  min_lr: 0.000214  loss: 3.8684 (3.6069)  weight_decay: 0.0500 (0.0500)  time: 0.4851  data: 0.1333  max mem: 21002
Epoch: [258]  [150/312]  eta: 0:01:36  lr: 0.000214  min_lr: 0.000214  loss: 3.5750 (3.5965)  weight_decay: 0.0500 (0.0500)  time: 0.4887  data: 0.1561  max mem: 21002
Epoch: [258]  [160/312]  eta: 0:01:31  lr: 0.000214  min_lr: 0.000214  loss: 3.5649 (3.5972)  weight_decay: 0.0500 (0.0500)  time: 0.6857  data: 0.2628  max mem: 21002
Epoch: [258]  [170/312]  eta: 0:01:22  lr: 0.000213  min_lr: 0.000213  loss: 3.7522 (3.6014)  weight_decay: 0.0500 (0.0500)  time: 0.4892  data: 0.1161  max mem: 21002
Epoch: [258]  [180/312]  eta: 0:01:17  lr: 0.000213  min_lr: 0.000213  loss: 3.8082 (3.5975)  weight_decay: 0.0500 (0.0500)  time: 0.4765  data: 0.1331  max mem: 21002
Epoch: [258]  [190/312]  eta: 0:01:09  lr: 0.000213  min_lr: 0.000213  loss: 3.8082 (3.5996)  weight_decay: 0.0500 (0.0500)  time: 0.4757  data: 0.1296  max mem: 21002
Epoch: [258]  [200/312]  eta: 0:01:04  lr: 0.000212  min_lr: 0.000212  loss: 3.7705 (3.5945)  weight_decay: 0.0500 (0.0500)  time: 0.4839  data: 0.1113  max mem: 21002
Epoch: [258]  [210/312]  eta: 0:00:59  lr: 0.000212  min_lr: 0.000212  loss: 3.5172 (3.5978)  weight_decay: 0.0500 (0.0500)  time: 0.6976  data: 0.2186  max mem: 21002
Epoch: [258]  [220/312]  eta: 0:00:52  lr: 0.000212  min_lr: 0.000212  loss: 3.8286 (3.6011)  weight_decay: 0.0500 (0.0500)  time: 0.5016  data: 0.1080  max mem: 21002
Epoch: [258]  [230/312]  eta: 0:00:47  lr: 0.000211  min_lr: 0.000211  loss: 3.8523 (3.6043)  weight_decay: 0.0500 (0.0500)  time: 0.4789  data: 0.0992  max mem: 21002
Epoch: [258]  [240/312]  eta: 0:00:41  lr: 0.000211  min_lr: 0.000211  loss: 3.8523 (3.6141)  weight_decay: 0.0500 (0.0500)  time: 0.6719  data: 0.2004  max mem: 21002
Epoch: [258]  [250/312]  eta: 0:00:35  lr: 0.000211  min_lr: 0.000211  loss: 3.6747 (3.6062)  weight_decay: 0.0500 (0.0500)  time: 0.4848  data: 0.1027  max mem: 21002
Epoch: [258]  [260/312]  eta: 0:00:29  lr: 0.000211  min_lr: 0.000211  loss: 3.4706 (3.6036)  weight_decay: 0.0500 (0.0500)  time: 0.4889  data: 0.1041  max mem: 21002
Epoch: [258]  [270/312]  eta: 0:00:23  lr: 0.000210  min_lr: 0.000210  loss: 3.6642 (3.6022)  weight_decay: 0.0500 (0.0500)  time: 0.4983  data: 0.1042  max mem: 21002
Epoch: [258]  [280/312]  eta: 0:00:18  lr: 0.000210  min_lr: 0.000210  loss: 3.4614 (3.5917)  weight_decay: 0.0500 (0.0500)  time: 0.4898  data: 0.0838  max mem: 21002
Epoch: [258]  [290/312]  eta: 0:00:12  lr: 0.000210  min_lr: 0.000210  loss: 3.3316 (3.5889)  weight_decay: 0.0500 (0.0500)  time: 0.6228  data: 0.1457  max mem: 21002
Epoch: [258]  [300/312]  eta: 0:00:06  lr: 0.000209  min_lr: 0.000209  loss: 3.6578 (3.5858)  weight_decay: 0.0500 (0.0500)  time: 0.4305  data: 0.0633  max mem: 21002
Epoch: [258]  [310/312]  eta: 0:00:01  lr: 0.000209  min_lr: 0.000209  loss: 3.8570 (3.5958)  weight_decay: 0.0500 (0.0500)  time: 0.2775  data: 0.0001  max mem: 21002
Epoch: [258]  [311/312]  eta: 0:00:00  lr: 0.000209  min_lr: 0.000209  loss: 3.7281 (3.5927)  weight_decay: 0.0500 (0.0500)  time: 0.2774  data: 0.0001  max mem: 21002
Epoch: [258] Total time: 0:02:51 (0.5502 s / it)
Averaged stats: lr: 0.000209  min_lr: 0.000209  loss: 3.7281 (3.6199)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.1557 (1.1557)  acc1: 80.5990 (80.5990)  acc5: 95.1823 (95.1823)  time: 8.8161  data: 8.6973  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4608 (1.3574)  acc1: 70.9635 (72.4800)  acc5: 91.5365 (91.4880)  time: 1.0657  data: 0.9664  max mem: 21002
Test: Total time: 0:00:09 (1.0820 s / it)
* Acc@1 72.498 Acc@5 91.404 loss 1.362
Accuracy of the model on the 50000 test images: 72.5%
Max accuracy: 72.69%
Epoch: [259]  [  0/312]  eta: 1:01:58  lr: 0.000209  min_lr: 0.000209  loss: 3.5129 (3.5129)  weight_decay: 0.0500 (0.0500)  time: 11.9183  data: 8.4844  max mem: 21002
Epoch: [259]  [ 10/312]  eta: 0:07:53  lr: 0.000209  min_lr: 0.000209  loss: 3.8294 (3.5907)  weight_decay: 0.0500 (0.0500)  time: 1.5678  data: 0.9945  max mem: 21002
Epoch: [259]  [ 20/312]  eta: 0:05:10  lr: 0.000208  min_lr: 0.000208  loss: 3.8845 (3.6990)  weight_decay: 0.0500 (0.0500)  time: 0.5209  data: 0.1668  max mem: 21002
Epoch: [259]  [ 30/312]  eta: 0:03:49  lr: 0.000208  min_lr: 0.000208  loss: 3.7203 (3.5612)  weight_decay: 0.0500 (0.0500)  time: 0.3992  data: 0.0443  max mem: 21002
Epoch: [259]  [ 40/312]  eta: 0:03:20  lr: 0.000208  min_lr: 0.000208  loss: 3.4481 (3.5736)  weight_decay: 0.0500 (0.0500)  time: 0.3970  data: 0.0466  max mem: 21002
Epoch: [259]  [ 50/312]  eta: 0:03:10  lr: 0.000207  min_lr: 0.000207  loss: 3.4481 (3.5559)  weight_decay: 0.0500 (0.0500)  time: 0.5899  data: 0.1102  max mem: 21002
Epoch: [259]  [ 60/312]  eta: 0:02:48  lr: 0.000207  min_lr: 0.000207  loss: 3.6700 (3.5712)  weight_decay: 0.0500 (0.0500)  time: 0.5230  data: 0.1070  max mem: 21002
Epoch: [259]  [ 70/312]  eta: 0:02:40  lr: 0.000207  min_lr: 0.000207  loss: 3.7446 (3.5844)  weight_decay: 0.0500 (0.0500)  time: 0.5075  data: 0.1161  max mem: 21002
Epoch: [259]  [ 80/312]  eta: 0:02:29  lr: 0.000206  min_lr: 0.000206  loss: 3.5882 (3.5600)  weight_decay: 0.0500 (0.0500)  time: 0.5711  data: 0.0751  max mem: 21002
Epoch: [259]  [ 90/312]  eta: 0:02:19  lr: 0.000206  min_lr: 0.000206  loss: 3.5729 (3.5663)  weight_decay: 0.0500 (0.0500)  time: 0.5083  data: 0.1215  max mem: 21002
Epoch: [259]  [100/312]  eta: 0:02:12  lr: 0.000206  min_lr: 0.000206  loss: 3.5980 (3.5486)  weight_decay: 0.0500 (0.0500)  time: 0.5430  data: 0.1667  max mem: 21002
Epoch: [259]  [110/312]  eta: 0:02:00  lr: 0.000205  min_lr: 0.000205  loss: 3.8520 (3.5740)  weight_decay: 0.0500 (0.0500)  time: 0.4527  data: 0.0754  max mem: 21002
Epoch: [259]  [120/312]  eta: 0:01:58  lr: 0.000205  min_lr: 0.000205  loss: 3.8538 (3.5677)  weight_decay: 0.0500 (0.0500)  time: 0.5696  data: 0.1616  max mem: 21002
Epoch: [259]  [130/312]  eta: 0:01:49  lr: 0.000205  min_lr: 0.000205  loss: 3.7429 (3.5763)  weight_decay: 0.0500 (0.0500)  time: 0.6094  data: 0.1348  max mem: 21002
Epoch: [259]  [140/312]  eta: 0:01:40  lr: 0.000204  min_lr: 0.000204  loss: 3.7728 (3.5914)  weight_decay: 0.0500 (0.0500)  time: 0.4089  data: 0.0596  max mem: 21002
Epoch: [259]  [150/312]  eta: 0:01:37  lr: 0.000204  min_lr: 0.000204  loss: 3.6780 (3.6014)  weight_decay: 0.0500 (0.0500)  time: 0.6094  data: 0.1937  max mem: 21002
Epoch: [259]  [160/312]  eta: 0:01:29  lr: 0.000204  min_lr: 0.000204  loss: 3.5776 (3.5957)  weight_decay: 0.0500 (0.0500)  time: 0.6360  data: 0.1363  max mem: 21002
Epoch: [259]  [170/312]  eta: 0:01:23  lr: 0.000203  min_lr: 0.000203  loss: 3.5809 (3.5993)  weight_decay: 0.0500 (0.0500)  time: 0.4638  data: 0.0583  max mem: 21002
Epoch: [259]  [180/312]  eta: 0:01:16  lr: 0.000203  min_lr: 0.000203  loss: 3.8328 (3.6168)  weight_decay: 0.0500 (0.0500)  time: 0.4892  data: 0.0763  max mem: 21002
Epoch: [259]  [190/312]  eta: 0:01:09  lr: 0.000203  min_lr: 0.000203  loss: 3.8424 (3.6193)  weight_decay: 0.0500 (0.0500)  time: 0.4501  data: 0.0729  max mem: 21002
Epoch: [259]  [200/312]  eta: 0:01:04  lr: 0.000203  min_lr: 0.000203  loss: 3.4517 (3.5958)  weight_decay: 0.0500 (0.0500)  time: 0.5737  data: 0.1577  max mem: 21002
Epoch: [259]  [210/312]  eta: 0:00:58  lr: 0.000202  min_lr: 0.000202  loss: 3.6081 (3.5933)  weight_decay: 0.0500 (0.0500)  time: 0.6041  data: 0.1043  max mem: 21002
Epoch: [259]  [220/312]  eta: 0:00:52  lr: 0.000202  min_lr: 0.000202  loss: 3.7249 (3.6019)  weight_decay: 0.0500 (0.0500)  time: 0.4352  data: 0.0647  max mem: 21002
Epoch: [259]  [230/312]  eta: 0:00:47  lr: 0.000202  min_lr: 0.000202  loss: 3.8909 (3.6132)  weight_decay: 0.0500 (0.0500)  time: 0.5994  data: 0.1781  max mem: 21002
Epoch: [259]  [240/312]  eta: 0:00:41  lr: 0.000201  min_lr: 0.000201  loss: 3.8909 (3.6136)  weight_decay: 0.0500 (0.0500)  time: 0.6200  data: 0.1141  max mem: 21002
Epoch: [259]  [250/312]  eta: 0:00:35  lr: 0.000201  min_lr: 0.000201  loss: 3.7102 (3.6073)  weight_decay: 0.0500 (0.0500)  time: 0.4874  data: 0.0842  max mem: 21002
Epoch: [259]  [260/312]  eta: 0:00:29  lr: 0.000201  min_lr: 0.000201  loss: 3.8617 (3.6147)  weight_decay: 0.0500 (0.0500)  time: 0.5244  data: 0.0922  max mem: 21002
Epoch: [259]  [270/312]  eta: 0:00:23  lr: 0.000200  min_lr: 0.000200  loss: 3.8025 (3.6104)  weight_decay: 0.0500 (0.0500)  time: 0.4709  data: 0.0710  max mem: 21002
Epoch: [259]  [280/312]  eta: 0:00:18  lr: 0.000200  min_lr: 0.000200  loss: 3.6897 (3.6133)  weight_decay: 0.0500 (0.0500)  time: 0.5565  data: 0.1507  max mem: 21002
Epoch: [259]  [290/312]  eta: 0:00:12  lr: 0.000200  min_lr: 0.000200  loss: 3.8069 (3.6087)  weight_decay: 0.0500 (0.0500)  time: 0.5955  data: 0.0979  max mem: 21002
Epoch: [259]  [300/312]  eta: 0:00:06  lr: 0.000199  min_lr: 0.000199  loss: 3.7988 (3.6043)  weight_decay: 0.0500 (0.0500)  time: 0.4246  data: 0.0509  max mem: 21002
Epoch: [259]  [310/312]  eta: 0:00:01  lr: 0.000199  min_lr: 0.000199  loss: 3.6675 (3.6007)  weight_decay: 0.0500 (0.0500)  time: 0.3175  data: 0.0411  max mem: 21002
Epoch: [259]  [311/312]  eta: 0:00:00  lr: 0.000199  min_lr: 0.000199  loss: 2.9817 (3.5969)  weight_decay: 0.0500 (0.0500)  time: 0.3175  data: 0.0410  max mem: 21002
Epoch: [259] Total time: 0:02:51 (0.5513 s / it)
Averaged stats: lr: 0.000199  min_lr: 0.000199  loss: 2.9817 (3.6131)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:11  loss: 1.0294 (1.0294)  acc1: 80.5990 (80.5990)  acc5: 95.5729 (95.5729)  time: 7.9015  data: 7.7827  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3549 (1.2706)  acc1: 72.7865 (73.2320)  acc5: 91.1458 (91.3600)  time: 1.0477  data: 0.9512  max mem: 21002
Test: Total time: 0:00:09 (1.0631 s / it)
* Acc@1 72.744 Acc@5 91.528 loss 1.274
Accuracy of the model on the 50000 test images: 72.7%
Max accuracy: 72.74%
Epoch: [260]  [  0/312]  eta: 1:01:22  lr: 0.000199  min_lr: 0.000199  loss: 3.9244 (3.9244)  weight_decay: 0.0500 (0.0500)  time: 11.8026  data: 9.4712  max mem: 21002
Epoch: [260]  [ 10/312]  eta: 0:07:59  lr: 0.000199  min_lr: 0.000199  loss: 3.3915 (3.3692)  weight_decay: 0.0500 (0.0500)  time: 1.5886  data: 1.0098  max mem: 21002
Epoch: [260]  [ 20/312]  eta: 0:04:58  lr: 0.000198  min_lr: 0.000198  loss: 3.2349 (3.3600)  weight_decay: 0.0500 (0.0500)  time: 0.4821  data: 0.1130  max mem: 21002
Epoch: [260]  [ 30/312]  eta: 0:03:41  lr: 0.000198  min_lr: 0.000198  loss: 3.6899 (3.3995)  weight_decay: 0.0500 (0.0500)  time: 0.3421  data: 0.0314  max mem: 21002
Epoch: [260]  [ 40/312]  eta: 0:03:20  lr: 0.000198  min_lr: 0.000198  loss: 3.5949 (3.4147)  weight_decay: 0.0500 (0.0500)  time: 0.4362  data: 0.0580  max mem: 21002
Epoch: [260]  [ 50/312]  eta: 0:03:05  lr: 0.000197  min_lr: 0.000197  loss: 3.5949 (3.4749)  weight_decay: 0.0500 (0.0500)  time: 0.5875  data: 0.1515  max mem: 21002
Epoch: [260]  [ 60/312]  eta: 0:02:41  lr: 0.000197  min_lr: 0.000197  loss: 3.8542 (3.5370)  weight_decay: 0.0500 (0.0500)  time: 0.4476  data: 0.0942  max mem: 21002
Epoch: [260]  [ 70/312]  eta: 0:02:33  lr: 0.000197  min_lr: 0.000197  loss: 3.8542 (3.5452)  weight_decay: 0.0500 (0.0500)  time: 0.4515  data: 0.1299  max mem: 21002
Epoch: [260]  [ 80/312]  eta: 0:02:28  lr: 0.000197  min_lr: 0.000197  loss: 3.7471 (3.5632)  weight_decay: 0.0500 (0.0500)  time: 0.6324  data: 0.3194  max mem: 21002
Epoch: [260]  [ 90/312]  eta: 0:02:14  lr: 0.000196  min_lr: 0.000196  loss: 3.7471 (3.5604)  weight_decay: 0.0500 (0.0500)  time: 0.5026  data: 0.2138  max mem: 21002
Epoch: [260]  [100/312]  eta: 0:02:09  lr: 0.000196  min_lr: 0.000196  loss: 3.4746 (3.5276)  weight_decay: 0.0500 (0.0500)  time: 0.5033  data: 0.2146  max mem: 21002
Epoch: [260]  [110/312]  eta: 0:01:57  lr: 0.000196  min_lr: 0.000196  loss: 3.1231 (3.5324)  weight_decay: 0.0500 (0.0500)  time: 0.4772  data: 0.1909  max mem: 21002
Epoch: [260]  [120/312]  eta: 0:01:53  lr: 0.000195  min_lr: 0.000195  loss: 3.6699 (3.5418)  weight_decay: 0.0500 (0.0500)  time: 0.4979  data: 0.2096  max mem: 21002
Epoch: [260]  [130/312]  eta: 0:01:48  lr: 0.000195  min_lr: 0.000195  loss: 3.6934 (3.5540)  weight_decay: 0.0500 (0.0500)  time: 0.6782  data: 0.3826  max mem: 21002
Epoch: [260]  [140/312]  eta: 0:01:38  lr: 0.000195  min_lr: 0.000195  loss: 3.7018 (3.5527)  weight_decay: 0.0500 (0.0500)  time: 0.4676  data: 0.1737  max mem: 21002
Epoch: [260]  [150/312]  eta: 0:01:34  lr: 0.000194  min_lr: 0.000194  loss: 3.4679 (3.5624)  weight_decay: 0.0500 (0.0500)  time: 0.4752  data: 0.1862  max mem: 21002
Epoch: [260]  [160/312]  eta: 0:01:29  lr: 0.000194  min_lr: 0.000194  loss: 3.7239 (3.5779)  weight_decay: 0.0500 (0.0500)  time: 0.6789  data: 0.3699  max mem: 21002
Epoch: [260]  [170/312]  eta: 0:01:21  lr: 0.000194  min_lr: 0.000194  loss: 3.7715 (3.5839)  weight_decay: 0.0500 (0.0500)  time: 0.5399  data: 0.1843  max mem: 21002
Epoch: [260]  [180/312]  eta: 0:01:16  lr: 0.000193  min_lr: 0.000193  loss: 3.7750 (3.5983)  weight_decay: 0.0500 (0.0500)  time: 0.4819  data: 0.1210  max mem: 21002
Epoch: [260]  [190/312]  eta: 0:01:09  lr: 0.000193  min_lr: 0.000193  loss: 3.7288 (3.5996)  weight_decay: 0.0500 (0.0500)  time: 0.4986  data: 0.1210  max mem: 21002
Epoch: [260]  [200/312]  eta: 0:01:03  lr: 0.000193  min_lr: 0.000193  loss: 3.7288 (3.6072)  weight_decay: 0.0500 (0.0500)  time: 0.4947  data: 0.1076  max mem: 21002
Epoch: [260]  [210/312]  eta: 0:00:58  lr: 0.000193  min_lr: 0.000193  loss: 3.8835 (3.6205)  weight_decay: 0.0500 (0.0500)  time: 0.6620  data: 0.2296  max mem: 21002
Epoch: [260]  [220/312]  eta: 0:00:52  lr: 0.000192  min_lr: 0.000192  loss: 3.8886 (3.6257)  weight_decay: 0.0500 (0.0500)  time: 0.5436  data: 0.1229  max mem: 21002
Epoch: [260]  [230/312]  eta: 0:00:46  lr: 0.000192  min_lr: 0.000192  loss: 3.7520 (3.6254)  weight_decay: 0.0500 (0.0500)  time: 0.4788  data: 0.1218  max mem: 21002
Epoch: [260]  [240/312]  eta: 0:00:40  lr: 0.000192  min_lr: 0.000192  loss: 3.5846 (3.6287)  weight_decay: 0.0500 (0.0500)  time: 0.5969  data: 0.1924  max mem: 21002
Epoch: [260]  [250/312]  eta: 0:00:35  lr: 0.000191  min_lr: 0.000191  loss: 3.5733 (3.6193)  weight_decay: 0.0500 (0.0500)  time: 0.5239  data: 0.1364  max mem: 21002
Epoch: [260]  [260/312]  eta: 0:00:29  lr: 0.000191  min_lr: 0.000191  loss: 3.5733 (3.6141)  weight_decay: 0.0500 (0.0500)  time: 0.5949  data: 0.2136  max mem: 21002
Epoch: [260]  [270/312]  eta: 0:00:23  lr: 0.000191  min_lr: 0.000191  loss: 3.8088 (3.6168)  weight_decay: 0.0500 (0.0500)  time: 0.5000  data: 0.1487  max mem: 21002
Epoch: [260]  [280/312]  eta: 0:00:18  lr: 0.000190  min_lr: 0.000190  loss: 3.9226 (3.6278)  weight_decay: 0.0500 (0.0500)  time: 0.5145  data: 0.1325  max mem: 21002
Epoch: [260]  [290/312]  eta: 0:00:12  lr: 0.000190  min_lr: 0.000190  loss: 3.8318 (3.6248)  weight_decay: 0.0500 (0.0500)  time: 0.6428  data: 0.1932  max mem: 21002
Epoch: [260]  [300/312]  eta: 0:00:06  lr: 0.000190  min_lr: 0.000190  loss: 3.7376 (3.6257)  weight_decay: 0.0500 (0.0500)  time: 0.4103  data: 0.0611  max mem: 21002
Epoch: [260]  [310/312]  eta: 0:00:01  lr: 0.000189  min_lr: 0.000189  loss: 3.6980 (3.6215)  weight_decay: 0.0500 (0.0500)  time: 0.2770  data: 0.0001  max mem: 21002
Epoch: [260]  [311/312]  eta: 0:00:00  lr: 0.000189  min_lr: 0.000189  loss: 3.6980 (3.6220)  weight_decay: 0.0500 (0.0500)  time: 0.2769  data: 0.0001  max mem: 21002
Epoch: [260] Total time: 0:02:51 (0.5503 s / it)
Averaged stats: lr: 0.000189  min_lr: 0.000189  loss: 3.6980 (3.6141)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.1198 (1.1198)  acc1: 81.1198 (81.1198)  acc5: 94.9219 (94.9219)  time: 8.2469  data: 8.1282  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4410 (1.3684)  acc1: 72.5260 (73.0560)  acc5: 90.6250 (91.5200)  time: 1.0069  data: 0.9081  max mem: 21002
Test: Total time: 0:00:09 (1.0171 s / it)
* Acc@1 72.868 Acc@5 91.462 loss 1.375
Accuracy of the model on the 50000 test images: 72.9%
Max accuracy: 72.87%
Epoch: [261]  [  0/312]  eta: 1:01:14  lr: 0.000189  min_lr: 0.000189  loss: 3.6210 (3.6210)  weight_decay: 0.0500 (0.0500)  time: 11.7785  data: 9.9741  max mem: 21002
Epoch: [261]  [ 10/312]  eta: 0:07:45  lr: 0.000189  min_lr: 0.000189  loss: 3.9695 (3.7442)  weight_decay: 0.0500 (0.0500)  time: 1.5428  data: 0.9632  max mem: 21002
Epoch: [261]  [ 20/312]  eta: 0:05:09  lr: 0.000189  min_lr: 0.000189  loss: 3.7192 (3.5796)  weight_decay: 0.0500 (0.0500)  time: 0.5239  data: 0.1330  max mem: 21002
Epoch: [261]  [ 30/312]  eta: 0:03:49  lr: 0.000188  min_lr: 0.000188  loss: 3.5523 (3.5886)  weight_decay: 0.0500 (0.0500)  time: 0.4111  data: 0.1023  max mem: 21002
Epoch: [261]  [ 40/312]  eta: 0:03:26  lr: 0.000188  min_lr: 0.000188  loss: 3.5715 (3.5835)  weight_decay: 0.0500 (0.0500)  time: 0.4428  data: 0.1074  max mem: 21002
Epoch: [261]  [ 50/312]  eta: 0:03:14  lr: 0.000188  min_lr: 0.000188  loss: 3.6407 (3.5753)  weight_decay: 0.0500 (0.0500)  time: 0.6300  data: 0.2531  max mem: 21002
Epoch: [261]  [ 60/312]  eta: 0:02:48  lr: 0.000188  min_lr: 0.000188  loss: 3.7625 (3.5887)  weight_decay: 0.0500 (0.0500)  time: 0.4781  data: 0.1465  max mem: 21002
Epoch: [261]  [ 70/312]  eta: 0:02:39  lr: 0.000187  min_lr: 0.000187  loss: 3.5148 (3.5680)  weight_decay: 0.0500 (0.0500)  time: 0.4473  data: 0.1583  max mem: 21002
Epoch: [261]  [ 80/312]  eta: 0:02:31  lr: 0.000187  min_lr: 0.000187  loss: 3.3462 (3.5409)  weight_decay: 0.0500 (0.0500)  time: 0.6174  data: 0.3292  max mem: 21002
Epoch: [261]  [ 90/312]  eta: 0:02:16  lr: 0.000187  min_lr: 0.000187  loss: 3.4166 (3.5380)  weight_decay: 0.0500 (0.0500)  time: 0.4573  data: 0.1716  max mem: 21002
Epoch: [261]  [100/312]  eta: 0:02:11  lr: 0.000186  min_lr: 0.000186  loss: 3.6268 (3.5428)  weight_decay: 0.0500 (0.0500)  time: 0.4724  data: 0.1860  max mem: 21002
Epoch: [261]  [110/312]  eta: 0:01:58  lr: 0.000186  min_lr: 0.000186  loss: 3.5335 (3.5390)  weight_decay: 0.0500 (0.0500)  time: 0.4713  data: 0.1860  max mem: 21002
Epoch: [261]  [120/312]  eta: 0:01:54  lr: 0.000186  min_lr: 0.000186  loss: 3.6576 (3.5575)  weight_decay: 0.0500 (0.0500)  time: 0.4814  data: 0.1973  max mem: 21002
Epoch: [261]  [130/312]  eta: 0:01:49  lr: 0.000185  min_lr: 0.000185  loss: 3.8423 (3.5550)  weight_decay: 0.0500 (0.0500)  time: 0.6851  data: 0.4009  max mem: 21002
Epoch: [261]  [140/312]  eta: 0:01:39  lr: 0.000185  min_lr: 0.000185  loss: 3.3750 (3.5377)  weight_decay: 0.0500 (0.0500)  time: 0.4886  data: 0.2042  max mem: 21002
Epoch: [261]  [150/312]  eta: 0:01:35  lr: 0.000185  min_lr: 0.000185  loss: 3.5806 (3.5571)  weight_decay: 0.0500 (0.0500)  time: 0.4770  data: 0.1912  max mem: 21002
Epoch: [261]  [160/312]  eta: 0:01:29  lr: 0.000185  min_lr: 0.000185  loss: 3.9886 (3.5753)  weight_decay: 0.0500 (0.0500)  time: 0.6220  data: 0.3355  max mem: 21002
Epoch: [261]  [170/312]  eta: 0:01:21  lr: 0.000184  min_lr: 0.000184  loss: 3.9031 (3.5721)  weight_decay: 0.0500 (0.0500)  time: 0.4759  data: 0.1617  max mem: 21002
Epoch: [261]  [180/312]  eta: 0:01:16  lr: 0.000184  min_lr: 0.000184  loss: 3.8885 (3.5767)  weight_decay: 0.0500 (0.0500)  time: 0.5232  data: 0.1664  max mem: 21002
Epoch: [261]  [190/312]  eta: 0:01:08  lr: 0.000184  min_lr: 0.000184  loss: 3.8885 (3.5785)  weight_decay: 0.0500 (0.0500)  time: 0.4786  data: 0.1494  max mem: 21002
Epoch: [261]  [200/312]  eta: 0:01:03  lr: 0.000183  min_lr: 0.000183  loss: 3.6065 (3.5734)  weight_decay: 0.0500 (0.0500)  time: 0.4729  data: 0.1635  max mem: 21002
Epoch: [261]  [210/312]  eta: 0:00:58  lr: 0.000183  min_lr: 0.000183  loss: 3.5215 (3.5781)  weight_decay: 0.0500 (0.0500)  time: 0.6420  data: 0.3200  max mem: 21002
Epoch: [261]  [220/312]  eta: 0:00:51  lr: 0.000183  min_lr: 0.000183  loss: 3.7299 (3.5843)  weight_decay: 0.0500 (0.0500)  time: 0.4866  data: 0.1571  max mem: 21002
Epoch: [261]  [230/312]  eta: 0:00:46  lr: 0.000182  min_lr: 0.000182  loss: 3.7088 (3.5858)  weight_decay: 0.0500 (0.0500)  time: 0.5485  data: 0.1586  max mem: 21002
Epoch: [261]  [240/312]  eta: 0:00:40  lr: 0.000182  min_lr: 0.000182  loss: 3.7088 (3.5869)  weight_decay: 0.0500 (0.0500)  time: 0.6307  data: 0.2087  max mem: 21002
Epoch: [261]  [250/312]  eta: 0:00:34  lr: 0.000182  min_lr: 0.000182  loss: 3.7877 (3.5977)  weight_decay: 0.0500 (0.0500)  time: 0.4784  data: 0.0933  max mem: 21002
Epoch: [261]  [260/312]  eta: 0:00:29  lr: 0.000182  min_lr: 0.000182  loss: 3.7923 (3.5960)  weight_decay: 0.0500 (0.0500)  time: 0.5705  data: 0.1539  max mem: 21002
Epoch: [261]  [270/312]  eta: 0:00:23  lr: 0.000181  min_lr: 0.000181  loss: 3.8551 (3.6021)  weight_decay: 0.0500 (0.0500)  time: 0.4916  data: 0.1114  max mem: 21002
Epoch: [261]  [280/312]  eta: 0:00:18  lr: 0.000181  min_lr: 0.000181  loss: 3.7030 (3.5999)  weight_decay: 0.0500 (0.0500)  time: 0.5093  data: 0.0619  max mem: 21002
Epoch: [261]  [290/312]  eta: 0:00:12  lr: 0.000181  min_lr: 0.000181  loss: 3.7040 (3.6014)  weight_decay: 0.0500 (0.0500)  time: 0.6068  data: 0.0894  max mem: 21002
Epoch: [261]  [300/312]  eta: 0:00:06  lr: 0.000180  min_lr: 0.000180  loss: 3.8969 (3.6097)  weight_decay: 0.0500 (0.0500)  time: 0.3880  data: 0.0279  max mem: 21002
Epoch: [261]  [310/312]  eta: 0:00:01  lr: 0.000180  min_lr: 0.000180  loss: 3.8243 (3.6042)  weight_decay: 0.0500 (0.0500)  time: 0.2914  data: 0.0001  max mem: 21002
Epoch: [261]  [311/312]  eta: 0:00:00  lr: 0.000180  min_lr: 0.000180  loss: 3.8357 (3.6069)  weight_decay: 0.0500 (0.0500)  time: 0.2831  data: 0.0001  max mem: 21002
Epoch: [261] Total time: 0:02:50 (0.5456 s / it)
Averaged stats: lr: 0.000180  min_lr: 0.000180  loss: 3.8357 (3.6201)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.0934 (1.0934)  acc1: 81.6406 (81.6406)  acc5: 95.0521 (95.0521)  time: 8.7491  data: 8.6188  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4351 (1.3400)  acc1: 72.1354 (72.9760)  acc5: 91.5365 (91.5360)  time: 1.0623  data: 0.9577  max mem: 21002
Test: Total time: 0:00:09 (1.0791 s / it)
* Acc@1 72.728 Acc@5 91.606 loss 1.342
Accuracy of the model on the 50000 test images: 72.7%
Max accuracy: 72.87%
Epoch: [262]  [  0/312]  eta: 1:02:46  lr: 0.000180  min_lr: 0.000180  loss: 4.2683 (4.2683)  weight_decay: 0.0500 (0.0500)  time: 12.0720  data: 8.4644  max mem: 21002
Epoch: [262]  [ 10/312]  eta: 0:08:07  lr: 0.000180  min_lr: 0.000180  loss: 2.9971 (3.2840)  weight_decay: 0.0500 (0.0500)  time: 1.6139  data: 0.9699  max mem: 21002
Epoch: [262]  [ 20/312]  eta: 0:05:16  lr: 0.000179  min_lr: 0.000179  loss: 3.3845 (3.4533)  weight_decay: 0.0500 (0.0500)  time: 0.5334  data: 0.1566  max mem: 21002
Epoch: [262]  [ 30/312]  eta: 0:03:53  lr: 0.000179  min_lr: 0.000179  loss: 3.7731 (3.5553)  weight_decay: 0.0500 (0.0500)  time: 0.3950  data: 0.0467  max mem: 21002
Epoch: [262]  [ 40/312]  eta: 0:03:19  lr: 0.000179  min_lr: 0.000179  loss: 3.6106 (3.5174)  weight_decay: 0.0500 (0.0500)  time: 0.3677  data: 0.0071  max mem: 21002
Epoch: [262]  [ 50/312]  eta: 0:03:09  lr: 0.000179  min_lr: 0.000179  loss: 3.6647 (3.5325)  weight_decay: 0.0500 (0.0500)  time: 0.5663  data: 0.0739  max mem: 21002
Epoch: [262]  [ 60/312]  eta: 0:02:44  lr: 0.000178  min_lr: 0.000178  loss: 3.6647 (3.5356)  weight_decay: 0.0500 (0.0500)  time: 0.4864  data: 0.0675  max mem: 21002
Epoch: [262]  [ 70/312]  eta: 0:02:40  lr: 0.000178  min_lr: 0.000178  loss: 3.8118 (3.6007)  weight_decay: 0.0500 (0.0500)  time: 0.5084  data: 0.0719  max mem: 21002
Epoch: [262]  [ 80/312]  eta: 0:02:30  lr: 0.000178  min_lr: 0.000178  loss: 3.8697 (3.6095)  weight_decay: 0.0500 (0.0500)  time: 0.6364  data: 0.0718  max mem: 21002
Epoch: [262]  [ 90/312]  eta: 0:02:17  lr: 0.000177  min_lr: 0.000177  loss: 3.7179 (3.6142)  weight_decay: 0.0500 (0.0500)  time: 0.4661  data: 0.0555  max mem: 21002
Epoch: [262]  [100/312]  eta: 0:02:13  lr: 0.000177  min_lr: 0.000177  loss: 3.7009 (3.5888)  weight_decay: 0.0500 (0.0500)  time: 0.5609  data: 0.1389  max mem: 21002
Epoch: [262]  [110/312]  eta: 0:02:01  lr: 0.000177  min_lr: 0.000177  loss: 3.5678 (3.5928)  weight_decay: 0.0500 (0.0500)  time: 0.5076  data: 0.0840  max mem: 21002
Epoch: [262]  [120/312]  eta: 0:01:55  lr: 0.000176  min_lr: 0.000176  loss: 3.6740 (3.6151)  weight_decay: 0.0500 (0.0500)  time: 0.4668  data: 0.0710  max mem: 21002
Epoch: [262]  [130/312]  eta: 0:01:50  lr: 0.000176  min_lr: 0.000176  loss: 3.8175 (3.6136)  weight_decay: 0.0500 (0.0500)  time: 0.6476  data: 0.1325  max mem: 21002
Epoch: [262]  [140/312]  eta: 0:01:40  lr: 0.000176  min_lr: 0.000176  loss: 3.8578 (3.6376)  weight_decay: 0.0500 (0.0500)  time: 0.4787  data: 0.0730  max mem: 21002
Epoch: [262]  [150/312]  eta: 0:01:36  lr: 0.000176  min_lr: 0.000176  loss: 3.7757 (3.6202)  weight_decay: 0.0500 (0.0500)  time: 0.5169  data: 0.0992  max mem: 21002
Epoch: [262]  [160/312]  eta: 0:01:29  lr: 0.000175  min_lr: 0.000175  loss: 3.5904 (3.6287)  weight_decay: 0.0500 (0.0500)  time: 0.6168  data: 0.0884  max mem: 21002
Epoch: [262]  [170/312]  eta: 0:01:22  lr: 0.000175  min_lr: 0.000175  loss: 3.7056 (3.6258)  weight_decay: 0.0500 (0.0500)  time: 0.4481  data: 0.0541  max mem: 21002
Epoch: [262]  [180/312]  eta: 0:01:16  lr: 0.000175  min_lr: 0.000175  loss: 3.7781 (3.6282)  weight_decay: 0.0500 (0.0500)  time: 0.5187  data: 0.1256  max mem: 21002
Epoch: [262]  [190/312]  eta: 0:01:09  lr: 0.000174  min_lr: 0.000174  loss: 3.8345 (3.6287)  weight_decay: 0.0500 (0.0500)  time: 0.4823  data: 0.0875  max mem: 21002
Epoch: [262]  [200/312]  eta: 0:01:04  lr: 0.000174  min_lr: 0.000174  loss: 3.7133 (3.6283)  weight_decay: 0.0500 (0.0500)  time: 0.4796  data: 0.1114  max mem: 21002
Epoch: [262]  [210/312]  eta: 0:00:58  lr: 0.000174  min_lr: 0.000174  loss: 3.6405 (3.6137)  weight_decay: 0.0500 (0.0500)  time: 0.6185  data: 0.1757  max mem: 21002
Epoch: [262]  [220/312]  eta: 0:00:51  lr: 0.000174  min_lr: 0.000174  loss: 3.5532 (3.6126)  weight_decay: 0.0500 (0.0500)  time: 0.4899  data: 0.1095  max mem: 21002
Epoch: [262]  [230/312]  eta: 0:00:46  lr: 0.000173  min_lr: 0.000173  loss: 3.5532 (3.6118)  weight_decay: 0.0500 (0.0500)  time: 0.5397  data: 0.1177  max mem: 21002
Epoch: [262]  [240/312]  eta: 0:00:40  lr: 0.000173  min_lr: 0.000173  loss: 3.7290 (3.6188)  weight_decay: 0.0500 (0.0500)  time: 0.6142  data: 0.1148  max mem: 21002
Epoch: [262]  [250/312]  eta: 0:00:35  lr: 0.000173  min_lr: 0.000173  loss: 3.7248 (3.6207)  weight_decay: 0.0500 (0.0500)  time: 0.5209  data: 0.0931  max mem: 21002
Epoch: [262]  [260/312]  eta: 0:00:29  lr: 0.000172  min_lr: 0.000172  loss: 3.6833 (3.6186)  weight_decay: 0.0500 (0.0500)  time: 0.6058  data: 0.1221  max mem: 21002
Epoch: [262]  [270/312]  eta: 0:00:23  lr: 0.000172  min_lr: 0.000172  loss: 3.6833 (3.6149)  weight_decay: 0.0500 (0.0500)  time: 0.5034  data: 0.0564  max mem: 21002
Epoch: [262]  [280/312]  eta: 0:00:18  lr: 0.000172  min_lr: 0.000172  loss: 3.7648 (3.6232)  weight_decay: 0.0500 (0.0500)  time: 0.4819  data: 0.0381  max mem: 21002
Epoch: [262]  [290/312]  eta: 0:00:12  lr: 0.000171  min_lr: 0.000171  loss: 3.8164 (3.6256)  weight_decay: 0.0500 (0.0500)  time: 0.5260  data: 0.0777  max mem: 21002
Epoch: [262]  [300/312]  eta: 0:00:06  lr: 0.000171  min_lr: 0.000171  loss: 3.7039 (3.6290)  weight_decay: 0.0500 (0.0500)  time: 0.3770  data: 0.0406  max mem: 21002
Epoch: [262]  [310/312]  eta: 0:00:01  lr: 0.000171  min_lr: 0.000171  loss: 3.6682 (3.6238)  weight_decay: 0.0500 (0.0500)  time: 0.3219  data: 0.0001  max mem: 21002
Epoch: [262]  [311/312]  eta: 0:00:00  lr: 0.000171  min_lr: 0.000171  loss: 3.7032 (3.6244)  weight_decay: 0.0500 (0.0500)  time: 0.3191  data: 0.0001  max mem: 21002
Epoch: [262] Total time: 0:02:50 (0.5465 s / it)
Averaged stats: lr: 0.000171  min_lr: 0.000171  loss: 3.7032 (3.6166)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.0508 (1.0508)  acc1: 80.4688 (80.4688)  acc5: 95.1823 (95.1823)  time: 8.5805  data: 8.4617  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4247 (1.3278)  acc1: 71.3542 (72.5760)  acc5: 91.4062 (91.6480)  time: 1.0433  data: 0.9460  max mem: 21002
Test: Total time: 0:00:09 (1.0726 s / it)
* Acc@1 72.924 Acc@5 91.606 loss 1.332
Accuracy of the model on the 50000 test images: 72.9%
Max accuracy: 72.92%
Epoch: [263]  [  0/312]  eta: 1:03:16  lr: 0.000171  min_lr: 0.000171  loss: 2.3784 (2.3784)  weight_decay: 0.0500 (0.0500)  time: 12.1693  data: 9.9126  max mem: 21002
Epoch: [263]  [ 10/312]  eta: 0:08:25  lr: 0.000171  min_lr: 0.000171  loss: 3.8904 (3.6620)  weight_decay: 0.0500 (0.0500)  time: 1.6740  data: 1.0055  max mem: 21002
Epoch: [263]  [ 20/312]  eta: 0:04:58  lr: 0.000170  min_lr: 0.000170  loss: 3.7044 (3.5451)  weight_decay: 0.0500 (0.0500)  time: 0.4651  data: 0.0583  max mem: 21002
Epoch: [263]  [ 30/312]  eta: 0:03:41  lr: 0.000170  min_lr: 0.000170  loss: 3.5821 (3.5689)  weight_decay: 0.0500 (0.0500)  time: 0.2976  data: 0.0013  max mem: 21002
Epoch: [263]  [ 40/312]  eta: 0:03:06  lr: 0.000170  min_lr: 0.000170  loss: 3.5889 (3.5787)  weight_decay: 0.0500 (0.0500)  time: 0.3349  data: 0.0008  max mem: 21002
Epoch: [263]  [ 50/312]  eta: 0:02:54  lr: 0.000169  min_lr: 0.000169  loss: 3.1671 (3.5031)  weight_decay: 0.0500 (0.0500)  time: 0.4837  data: 0.1154  max mem: 21002
Epoch: [263]  [ 60/312]  eta: 0:02:38  lr: 0.000169  min_lr: 0.000169  loss: 3.5050 (3.5530)  weight_decay: 0.0500 (0.0500)  time: 0.5092  data: 0.1197  max mem: 21002
Epoch: [263]  [ 70/312]  eta: 0:02:35  lr: 0.000169  min_lr: 0.000169  loss: 3.7941 (3.5635)  weight_decay: 0.0500 (0.0500)  time: 0.5751  data: 0.0880  max mem: 21002
Epoch: [263]  [ 80/312]  eta: 0:02:23  lr: 0.000169  min_lr: 0.000169  loss: 3.7941 (3.5787)  weight_decay: 0.0500 (0.0500)  time: 0.5960  data: 0.1659  max mem: 21002
Epoch: [263]  [ 90/312]  eta: 0:02:16  lr: 0.000168  min_lr: 0.000168  loss: 3.5696 (3.5604)  weight_decay: 0.0500 (0.0500)  time: 0.5172  data: 0.0830  max mem: 21002
Epoch: [263]  [100/312]  eta: 0:02:12  lr: 0.000168  min_lr: 0.000168  loss: 3.5253 (3.5466)  weight_decay: 0.0500 (0.0500)  time: 0.6407  data: 0.0059  max mem: 21002
Epoch: [263]  [110/312]  eta: 0:02:01  lr: 0.000168  min_lr: 0.000168  loss: 3.4545 (3.5325)  weight_decay: 0.0500 (0.0500)  time: 0.5366  data: 0.0395  max mem: 21002
Epoch: [263]  [120/312]  eta: 0:01:56  lr: 0.000167  min_lr: 0.000167  loss: 3.1547 (3.5142)  weight_decay: 0.0500 (0.0500)  time: 0.5313  data: 0.0523  max mem: 21002
Epoch: [263]  [130/312]  eta: 0:01:46  lr: 0.000167  min_lr: 0.000167  loss: 3.4853 (3.5237)  weight_decay: 0.0500 (0.0500)  time: 0.5087  data: 0.0256  max mem: 21002
Epoch: [263]  [140/312]  eta: 0:01:41  lr: 0.000167  min_lr: 0.000167  loss: 3.7815 (3.5471)  weight_decay: 0.0500 (0.0500)  time: 0.4869  data: 0.0243  max mem: 21002
Epoch: [263]  [150/312]  eta: 0:01:36  lr: 0.000167  min_lr: 0.000167  loss: 3.7707 (3.5452)  weight_decay: 0.0500 (0.0500)  time: 0.6792  data: 0.0659  max mem: 21002
Epoch: [263]  [160/312]  eta: 0:01:28  lr: 0.000166  min_lr: 0.000166  loss: 3.5659 (3.5501)  weight_decay: 0.0500 (0.0500)  time: 0.4920  data: 0.0493  max mem: 21002
Epoch: [263]  [170/312]  eta: 0:01:22  lr: 0.000166  min_lr: 0.000166  loss: 3.7438 (3.5557)  weight_decay: 0.0500 (0.0500)  time: 0.4740  data: 0.0194  max mem: 21002
Epoch: [263]  [180/312]  eta: 0:01:17  lr: 0.000166  min_lr: 0.000166  loss: 3.8256 (3.5607)  weight_decay: 0.0500 (0.0500)  time: 0.6571  data: 0.0196  max mem: 21002
Epoch: [263]  [190/312]  eta: 0:01:09  lr: 0.000165  min_lr: 0.000165  loss: 3.7120 (3.5649)  weight_decay: 0.0500 (0.0500)  time: 0.4865  data: 0.0135  max mem: 21002
Epoch: [263]  [200/312]  eta: 0:01:04  lr: 0.000165  min_lr: 0.000165  loss: 3.8202 (3.5766)  weight_decay: 0.0500 (0.0500)  time: 0.4958  data: 0.0545  max mem: 21002
Epoch: [263]  [210/312]  eta: 0:00:58  lr: 0.000165  min_lr: 0.000165  loss: 3.8721 (3.5856)  weight_decay: 0.0500 (0.0500)  time: 0.5208  data: 0.0721  max mem: 21002
Epoch: [263]  [220/312]  eta: 0:00:52  lr: 0.000165  min_lr: 0.000165  loss: 3.7953 (3.5869)  weight_decay: 0.0500 (0.0500)  time: 0.4661  data: 0.0311  max mem: 21002
Epoch: [263]  [230/312]  eta: 0:00:47  lr: 0.000164  min_lr: 0.000164  loss: 3.5260 (3.5835)  weight_decay: 0.0500 (0.0500)  time: 0.6252  data: 0.0396  max mem: 21002
Epoch: [263]  [240/312]  eta: 0:00:40  lr: 0.000164  min_lr: 0.000164  loss: 3.8403 (3.5933)  weight_decay: 0.0500 (0.0500)  time: 0.5101  data: 0.0466  max mem: 21002
Epoch: [263]  [250/312]  eta: 0:00:35  lr: 0.000164  min_lr: 0.000164  loss: 3.8524 (3.5899)  weight_decay: 0.0500 (0.0500)  time: 0.4918  data: 0.0442  max mem: 21002
Epoch: [263]  [260/312]  eta: 0:00:29  lr: 0.000163  min_lr: 0.000163  loss: 3.6647 (3.5932)  weight_decay: 0.0500 (0.0500)  time: 0.6621  data: 0.0525  max mem: 21002
Epoch: [263]  [270/312]  eta: 0:00:23  lr: 0.000163  min_lr: 0.000163  loss: 3.5845 (3.5884)  weight_decay: 0.0500 (0.0500)  time: 0.4911  data: 0.0225  max mem: 21002
Epoch: [263]  [280/312]  eta: 0:00:18  lr: 0.000163  min_lr: 0.000163  loss: 3.4508 (3.5822)  weight_decay: 0.0500 (0.0500)  time: 0.4687  data: 0.0440  max mem: 21002
Epoch: [263]  [290/312]  eta: 0:00:12  lr: 0.000163  min_lr: 0.000163  loss: 3.7527 (3.5938)  weight_decay: 0.0500 (0.0500)  time: 0.5942  data: 0.0672  max mem: 21002
Epoch: [263]  [300/312]  eta: 0:00:06  lr: 0.000162  min_lr: 0.000162  loss: 3.8424 (3.5955)  weight_decay: 0.0500 (0.0500)  time: 0.4350  data: 0.0300  max mem: 21002
Epoch: [263]  [310/312]  eta: 0:00:01  lr: 0.000162  min_lr: 0.000162  loss: 3.6291 (3.5977)  weight_decay: 0.0500 (0.0500)  time: 0.2956  data: 0.0001  max mem: 21002
Epoch: [263]  [311/312]  eta: 0:00:00  lr: 0.000162  min_lr: 0.000162  loss: 3.6291 (3.5987)  weight_decay: 0.0500 (0.0500)  time: 0.2955  data: 0.0001  max mem: 21002
Epoch: [263] Total time: 0:02:51 (0.5495 s / it)
Averaged stats: lr: 0.000162  min_lr: 0.000162  loss: 3.6291 (3.6011)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.0914 (1.0914)  acc1: 80.4688 (80.4688)  acc5: 95.4427 (95.4427)  time: 8.7826  data: 8.6639  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4245 (1.3336)  acc1: 72.3958 (73.4400)  acc5: 91.5365 (91.7120)  time: 1.0612  data: 0.9627  max mem: 21002
Test: Total time: 0:00:09 (1.0738 s / it)
* Acc@1 72.856 Acc@5 91.594 loss 1.345
Accuracy of the model on the 50000 test images: 72.9%
Max accuracy: 72.92%
Epoch: [264]  [  0/312]  eta: 1:02:55  lr: 0.000162  min_lr: 0.000162  loss: 3.1852 (3.1852)  weight_decay: 0.0500 (0.0500)  time: 12.0999  data: 10.8783  max mem: 21002
Epoch: [264]  [ 10/312]  eta: 0:07:45  lr: 0.000162  min_lr: 0.000162  loss: 4.2198 (3.7921)  weight_decay: 0.0500 (0.0500)  time: 1.5407  data: 0.9899  max mem: 21002
Epoch: [264]  [ 20/312]  eta: 0:05:06  lr: 0.000161  min_lr: 0.000161  loss: 3.8817 (3.7210)  weight_decay: 0.0500 (0.0500)  time: 0.4982  data: 0.0607  max mem: 21002
Epoch: [264]  [ 30/312]  eta: 0:03:47  lr: 0.000161  min_lr: 0.000161  loss: 3.7361 (3.6666)  weight_decay: 0.0500 (0.0500)  time: 0.4008  data: 0.0605  max mem: 21002
Epoch: [264]  [ 40/312]  eta: 0:03:21  lr: 0.000161  min_lr: 0.000161  loss: 3.5564 (3.6544)  weight_decay: 0.0500 (0.0500)  time: 0.4175  data: 0.0927  max mem: 21002
Epoch: [264]  [ 50/312]  eta: 0:03:08  lr: 0.000160  min_lr: 0.000160  loss: 3.7612 (3.6616)  weight_decay: 0.0500 (0.0500)  time: 0.5852  data: 0.1922  max mem: 21002
Epoch: [264]  [ 60/312]  eta: 0:02:43  lr: 0.000160  min_lr: 0.000160  loss: 3.7612 (3.6454)  weight_decay: 0.0500 (0.0500)  time: 0.4568  data: 0.1012  max mem: 21002
Epoch: [264]  [ 70/312]  eta: 0:02:41  lr: 0.000160  min_lr: 0.000160  loss: 3.8360 (3.6550)  weight_decay: 0.0500 (0.0500)  time: 0.5378  data: 0.1390  max mem: 21002
Epoch: [264]  [ 80/312]  eta: 0:02:28  lr: 0.000160  min_lr: 0.000160  loss: 3.7838 (3.6565)  weight_decay: 0.0500 (0.0500)  time: 0.6205  data: 0.1733  max mem: 21002
Epoch: [264]  [ 90/312]  eta: 0:02:18  lr: 0.000159  min_lr: 0.000159  loss: 3.9361 (3.7045)  weight_decay: 0.0500 (0.0500)  time: 0.4761  data: 0.0625  max mem: 21002
Epoch: [264]  [100/312]  eta: 0:02:14  lr: 0.000159  min_lr: 0.000159  loss: 3.9895 (3.7256)  weight_decay: 0.0500 (0.0500)  time: 0.6051  data: 0.1094  max mem: 21002
Epoch: [264]  [110/312]  eta: 0:02:01  lr: 0.000159  min_lr: 0.000159  loss: 3.8860 (3.7211)  weight_decay: 0.0500 (0.0500)  time: 0.5007  data: 0.0828  max mem: 21002
Epoch: [264]  [120/312]  eta: 0:01:57  lr: 0.000159  min_lr: 0.000159  loss: 3.6524 (3.7136)  weight_decay: 0.0500 (0.0500)  time: 0.4839  data: 0.0688  max mem: 21002
Epoch: [264]  [130/312]  eta: 0:01:51  lr: 0.000158  min_lr: 0.000158  loss: 3.8035 (3.7217)  weight_decay: 0.0500 (0.0500)  time: 0.6656  data: 0.0993  max mem: 21002
Epoch: [264]  [140/312]  eta: 0:01:41  lr: 0.000158  min_lr: 0.000158  loss: 3.3622 (3.6662)  weight_decay: 0.0500 (0.0500)  time: 0.4740  data: 0.0350  max mem: 21002
Epoch: [264]  [150/312]  eta: 0:01:36  lr: 0.000158  min_lr: 0.000158  loss: 3.2479 (3.6589)  weight_decay: 0.0500 (0.0500)  time: 0.4810  data: 0.0652  max mem: 21002
Epoch: [264]  [160/312]  eta: 0:01:28  lr: 0.000157  min_lr: 0.000157  loss: 3.5394 (3.6440)  weight_decay: 0.0500 (0.0500)  time: 0.5462  data: 0.0967  max mem: 21002
Epoch: [264]  [170/312]  eta: 0:01:22  lr: 0.000157  min_lr: 0.000157  loss: 3.6134 (3.6461)  weight_decay: 0.0500 (0.0500)  time: 0.4652  data: 0.0658  max mem: 21002
Epoch: [264]  [180/312]  eta: 0:01:16  lr: 0.000157  min_lr: 0.000157  loss: 3.7233 (3.6394)  weight_decay: 0.0500 (0.0500)  time: 0.5670  data: 0.1262  max mem: 21002
Epoch: [264]  [190/312]  eta: 0:01:09  lr: 0.000157  min_lr: 0.000157  loss: 3.5232 (3.6324)  weight_decay: 0.0500 (0.0500)  time: 0.4572  data: 0.0964  max mem: 21002
Epoch: [264]  [200/312]  eta: 0:01:04  lr: 0.000156  min_lr: 0.000156  loss: 3.7227 (3.6326)  weight_decay: 0.0500 (0.0500)  time: 0.4918  data: 0.1192  max mem: 21002
Epoch: [264]  [210/312]  eta: 0:00:58  lr: 0.000156  min_lr: 0.000156  loss: 3.8199 (3.6399)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.2359  max mem: 21002
Epoch: [264]  [220/312]  eta: 0:00:51  lr: 0.000156  min_lr: 0.000156  loss: 3.8731 (3.6456)  weight_decay: 0.0500 (0.0500)  time: 0.4785  data: 0.1172  max mem: 21002
Epoch: [264]  [230/312]  eta: 0:00:46  lr: 0.000155  min_lr: 0.000155  loss: 3.8479 (3.6435)  weight_decay: 0.0500 (0.0500)  time: 0.4935  data: 0.0949  max mem: 21002
Epoch: [264]  [240/312]  eta: 0:00:40  lr: 0.000155  min_lr: 0.000155  loss: 3.9186 (3.6513)  weight_decay: 0.0500 (0.0500)  time: 0.5706  data: 0.1591  max mem: 21002
Epoch: [264]  [250/312]  eta: 0:00:35  lr: 0.000155  min_lr: 0.000155  loss: 3.8391 (3.6402)  weight_decay: 0.0500 (0.0500)  time: 0.5135  data: 0.1104  max mem: 21002
Epoch: [264]  [260/312]  eta: 0:00:29  lr: 0.000155  min_lr: 0.000155  loss: 3.5068 (3.6341)  weight_decay: 0.0500 (0.0500)  time: 0.6290  data: 0.0861  max mem: 21002
Epoch: [264]  [270/312]  eta: 0:00:23  lr: 0.000154  min_lr: 0.000154  loss: 3.4479 (3.6257)  weight_decay: 0.0500 (0.0500)  time: 0.4848  data: 0.0447  max mem: 21002
Epoch: [264]  [280/312]  eta: 0:00:18  lr: 0.000154  min_lr: 0.000154  loss: 3.5470 (3.6270)  weight_decay: 0.0500 (0.0500)  time: 0.4907  data: 0.0662  max mem: 21002
Epoch: [264]  [290/312]  eta: 0:00:12  lr: 0.000154  min_lr: 0.000154  loss: 3.8505 (3.6226)  weight_decay: 0.0500 (0.0500)  time: 0.6059  data: 0.0854  max mem: 21002
Epoch: [264]  [300/312]  eta: 0:00:06  lr: 0.000154  min_lr: 0.000154  loss: 3.7903 (3.6269)  weight_decay: 0.0500 (0.0500)  time: 0.4172  data: 0.0381  max mem: 21002
Epoch: [264]  [310/312]  eta: 0:00:01  lr: 0.000153  min_lr: 0.000153  loss: 3.8934 (3.6370)  weight_decay: 0.0500 (0.0500)  time: 0.2911  data: 0.0144  max mem: 21002
Epoch: [264]  [311/312]  eta: 0:00:00  lr: 0.000153  min_lr: 0.000153  loss: 3.8924 (3.6378)  weight_decay: 0.0500 (0.0500)  time: 0.2910  data: 0.0144  max mem: 21002
Epoch: [264] Total time: 0:02:51 (0.5489 s / it)
Averaged stats: lr: 0.000153  min_lr: 0.000153  loss: 3.8924 (3.5976)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.1840 (1.1840)  acc1: 80.2083 (80.2083)  acc5: 95.1823 (95.1823)  time: 8.8237  data: 8.7050  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.5042 (1.4067)  acc1: 73.3073 (73.3760)  acc5: 90.8854 (91.3920)  time: 1.0740  data: 0.9673  max mem: 21002
Test: Total time: 0:00:09 (1.1044 s / it)
* Acc@1 72.894 Acc@5 91.658 loss 1.411
Accuracy of the model on the 50000 test images: 72.9%
Max accuracy: 72.92%
Epoch: [265]  [  0/312]  eta: 1:05:03  lr: 0.000153  min_lr: 0.000153  loss: 3.9088 (3.9088)  weight_decay: 0.0500 (0.0500)  time: 12.5104  data: 12.2171  max mem: 21002
Epoch: [265]  [ 10/312]  eta: 0:08:17  lr: 0.000153  min_lr: 0.000153  loss: 3.8821 (3.5967)  weight_decay: 0.0500 (0.0500)  time: 1.6472  data: 1.1112  max mem: 21002
Epoch: [265]  [ 20/312]  eta: 0:05:00  lr: 0.000153  min_lr: 0.000153  loss: 3.8848 (3.7710)  weight_decay: 0.0500 (0.0500)  time: 0.4561  data: 0.0228  max mem: 21002
Epoch: [265]  [ 30/312]  eta: 0:03:42  lr: 0.000152  min_lr: 0.000152  loss: 3.7172 (3.6437)  weight_decay: 0.0500 (0.0500)  time: 0.3195  data: 0.0229  max mem: 21002
Epoch: [265]  [ 40/312]  eta: 0:03:07  lr: 0.000152  min_lr: 0.000152  loss: 3.6780 (3.6756)  weight_decay: 0.0500 (0.0500)  time: 0.3323  data: 0.0196  max mem: 21002
Epoch: [265]  [ 50/312]  eta: 0:02:59  lr: 0.000152  min_lr: 0.000152  loss: 3.8151 (3.6741)  weight_decay: 0.0500 (0.0500)  time: 0.5196  data: 0.1536  max mem: 21002
Epoch: [265]  [ 60/312]  eta: 0:02:36  lr: 0.000152  min_lr: 0.000152  loss: 3.7430 (3.6496)  weight_decay: 0.0500 (0.0500)  time: 0.4760  data: 0.1349  max mem: 21002
Epoch: [265]  [ 70/312]  eta: 0:02:32  lr: 0.000151  min_lr: 0.000151  loss: 3.7430 (3.6497)  weight_decay: 0.0500 (0.0500)  time: 0.4948  data: 0.1485  max mem: 21002
Epoch: [265]  [ 80/312]  eta: 0:02:24  lr: 0.000151  min_lr: 0.000151  loss: 3.7273 (3.6546)  weight_decay: 0.0500 (0.0500)  time: 0.6325  data: 0.2277  max mem: 21002
Epoch: [265]  [ 90/312]  eta: 0:02:14  lr: 0.000151  min_lr: 0.000151  loss: 3.6851 (3.6397)  weight_decay: 0.0500 (0.0500)  time: 0.5089  data: 0.1100  max mem: 21002
Epoch: [265]  [100/312]  eta: 0:02:09  lr: 0.000150  min_lr: 0.000150  loss: 3.4387 (3.6088)  weight_decay: 0.0500 (0.0500)  time: 0.5558  data: 0.0864  max mem: 21002
Epoch: [265]  [110/312]  eta: 0:01:58  lr: 0.000150  min_lr: 0.000150  loss: 3.3358 (3.5972)  weight_decay: 0.0500 (0.0500)  time: 0.4977  data: 0.0826  max mem: 21002
Epoch: [265]  [120/312]  eta: 0:01:54  lr: 0.000150  min_lr: 0.000150  loss: 3.5990 (3.5928)  weight_decay: 0.0500 (0.0500)  time: 0.5245  data: 0.0795  max mem: 21002
Epoch: [265]  [130/312]  eta: 0:01:49  lr: 0.000150  min_lr: 0.000150  loss: 3.6845 (3.6015)  weight_decay: 0.0500 (0.0500)  time: 0.6864  data: 0.0860  max mem: 21002
Epoch: [265]  [140/312]  eta: 0:01:39  lr: 0.000149  min_lr: 0.000149  loss: 3.6886 (3.6058)  weight_decay: 0.0500 (0.0500)  time: 0.4805  data: 0.0376  max mem: 21002
Epoch: [265]  [150/312]  eta: 0:01:36  lr: 0.000149  min_lr: 0.000149  loss: 3.6302 (3.5902)  weight_decay: 0.0500 (0.0500)  time: 0.5418  data: 0.0551  max mem: 21002
Epoch: [265]  [160/312]  eta: 0:01:28  lr: 0.000149  min_lr: 0.000149  loss: 3.0942 (3.5600)  weight_decay: 0.0500 (0.0500)  time: 0.5860  data: 0.0688  max mem: 21002
Epoch: [265]  [170/312]  eta: 0:01:22  lr: 0.000149  min_lr: 0.000149  loss: 3.2970 (3.5645)  weight_decay: 0.0500 (0.0500)  time: 0.4749  data: 0.0188  max mem: 21002
Epoch: [265]  [180/312]  eta: 0:01:17  lr: 0.000148  min_lr: 0.000148  loss: 3.6498 (3.5650)  weight_decay: 0.0500 (0.0500)  time: 0.6392  data: 0.0059  max mem: 21002
Epoch: [265]  [190/312]  eta: 0:01:09  lr: 0.000148  min_lr: 0.000148  loss: 3.5223 (3.5626)  weight_decay: 0.0500 (0.0500)  time: 0.5115  data: 0.0122  max mem: 21002
Epoch: [265]  [200/312]  eta: 0:01:05  lr: 0.000148  min_lr: 0.000148  loss: 3.4987 (3.5595)  weight_decay: 0.0500 (0.0500)  time: 0.5283  data: 0.0200  max mem: 21002
Epoch: [265]  [210/312]  eta: 0:00:59  lr: 0.000147  min_lr: 0.000147  loss: 3.7823 (3.5715)  weight_decay: 0.0500 (0.0500)  time: 0.7182  data: 0.0375  max mem: 21002
Epoch: [265]  [220/312]  eta: 0:00:52  lr: 0.000147  min_lr: 0.000147  loss: 3.9039 (3.5770)  weight_decay: 0.0500 (0.0500)  time: 0.5044  data: 0.0426  max mem: 21002
Epoch: [265]  [230/312]  eta: 0:00:47  lr: 0.000147  min_lr: 0.000147  loss: 3.6411 (3.5641)  weight_decay: 0.0500 (0.0500)  time: 0.5191  data: 0.0567  max mem: 21002
Epoch: [265]  [240/312]  eta: 0:00:41  lr: 0.000147  min_lr: 0.000147  loss: 3.6411 (3.5744)  weight_decay: 0.0500 (0.0500)  time: 0.5316  data: 0.0686  max mem: 21002
Epoch: [265]  [250/312]  eta: 0:00:35  lr: 0.000146  min_lr: 0.000146  loss: 3.8825 (3.5868)  weight_decay: 0.0500 (0.0500)  time: 0.4981  data: 0.0494  max mem: 21002
Epoch: [265]  [260/312]  eta: 0:00:29  lr: 0.000146  min_lr: 0.000146  loss: 3.8927 (3.6000)  weight_decay: 0.0500 (0.0500)  time: 0.6421  data: 0.0385  max mem: 21002
Epoch: [265]  [270/312]  eta: 0:00:23  lr: 0.000146  min_lr: 0.000146  loss: 3.8927 (3.6073)  weight_decay: 0.0500 (0.0500)  time: 0.4835  data: 0.0391  max mem: 21002
Epoch: [265]  [280/312]  eta: 0:00:18  lr: 0.000146  min_lr: 0.000146  loss: 3.8417 (3.6088)  weight_decay: 0.0500 (0.0500)  time: 0.4845  data: 0.0724  max mem: 21002
Epoch: [265]  [290/312]  eta: 0:00:12  lr: 0.000145  min_lr: 0.000145  loss: 3.7883 (3.6104)  weight_decay: 0.0500 (0.0500)  time: 0.5683  data: 0.0729  max mem: 21002
Epoch: [265]  [300/312]  eta: 0:00:06  lr: 0.000145  min_lr: 0.000145  loss: 3.7060 (3.6060)  weight_decay: 0.0500 (0.0500)  time: 0.3875  data: 0.0204  max mem: 21002
Epoch: [265]  [310/312]  eta: 0:00:01  lr: 0.000145  min_lr: 0.000145  loss: 3.5791 (3.6032)  weight_decay: 0.0500 (0.0500)  time: 0.2771  data: 0.0001  max mem: 21002
Epoch: [265]  [311/312]  eta: 0:00:00  lr: 0.000145  min_lr: 0.000145  loss: 3.5791 (3.6011)  weight_decay: 0.0500 (0.0500)  time: 0.2770  data: 0.0001  max mem: 21002
Epoch: [265] Total time: 0:02:51 (0.5512 s / it)
Averaged stats: lr: 0.000145  min_lr: 0.000145  loss: 3.5791 (3.6195)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 0.9893 (0.9893)  acc1: 81.3802 (81.3802)  acc5: 95.4427 (95.4427)  time: 8.9483  data: 8.8296  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3280 (1.2408)  acc1: 72.1354 (73.5040)  acc5: 91.7969 (91.9680)  time: 1.0832  data: 0.9811  max mem: 21002
Test: Total time: 0:00:10 (1.1202 s / it)
* Acc@1 73.426 Acc@5 91.856 loss 1.245
Accuracy of the model on the 50000 test images: 73.4%
Max accuracy: 73.43%
Epoch: [266]  [  0/312]  eta: 1:02:44  lr: 0.000145  min_lr: 0.000145  loss: 4.3335 (4.3335)  weight_decay: 0.0500 (0.0500)  time: 12.0670  data: 9.5352  max mem: 21002
Epoch: [266]  [ 10/312]  eta: 0:07:44  lr: 0.000144  min_lr: 0.000144  loss: 3.8490 (3.7454)  weight_decay: 0.0500 (0.0500)  time: 1.5364  data: 1.0466  max mem: 21002
Epoch: [266]  [ 20/312]  eta: 0:04:58  lr: 0.000144  min_lr: 0.000144  loss: 3.6954 (3.6629)  weight_decay: 0.0500 (0.0500)  time: 0.4716  data: 0.1878  max mem: 21002
Epoch: [266]  [ 30/312]  eta: 0:03:41  lr: 0.000144  min_lr: 0.000144  loss: 3.7705 (3.6773)  weight_decay: 0.0500 (0.0500)  time: 0.3747  data: 0.0892  max mem: 21002
Epoch: [266]  [ 40/312]  eta: 0:03:23  lr: 0.000144  min_lr: 0.000144  loss: 3.5796 (3.6497)  weight_decay: 0.0500 (0.0500)  time: 0.4550  data: 0.1614  max mem: 21002
Epoch: [266]  [ 50/312]  eta: 0:03:10  lr: 0.000143  min_lr: 0.000143  loss: 3.5276 (3.6285)  weight_decay: 0.0500 (0.0500)  time: 0.6357  data: 0.3442  max mem: 21002
Epoch: [266]  [ 60/312]  eta: 0:02:45  lr: 0.000143  min_lr: 0.000143  loss: 3.5276 (3.6160)  weight_decay: 0.0500 (0.0500)  time: 0.4682  data: 0.1834  max mem: 21002
Epoch: [266]  [ 70/312]  eta: 0:02:37  lr: 0.000143  min_lr: 0.000143  loss: 3.6854 (3.6061)  weight_decay: 0.0500 (0.0500)  time: 0.4565  data: 0.1702  max mem: 21002
Epoch: [266]  [ 80/312]  eta: 0:02:30  lr: 0.000143  min_lr: 0.000143  loss: 3.7909 (3.6283)  weight_decay: 0.0500 (0.0500)  time: 0.6299  data: 0.3450  max mem: 21002
Epoch: [266]  [ 90/312]  eta: 0:02:15  lr: 0.000142  min_lr: 0.000142  loss: 3.8219 (3.6412)  weight_decay: 0.0500 (0.0500)  time: 0.4682  data: 0.1826  max mem: 21002
Epoch: [266]  [100/312]  eta: 0:02:10  lr: 0.000142  min_lr: 0.000142  loss: 3.8119 (3.6224)  weight_decay: 0.0500 (0.0500)  time: 0.4900  data: 0.1671  max mem: 21002
Epoch: [266]  [110/312]  eta: 0:01:58  lr: 0.000142  min_lr: 0.000142  loss: 3.5849 (3.5999)  weight_decay: 0.0500 (0.0500)  time: 0.4802  data: 0.1606  max mem: 21002
Epoch: [266]  [120/312]  eta: 0:01:54  lr: 0.000142  min_lr: 0.000142  loss: 3.5627 (3.5978)  weight_decay: 0.0500 (0.0500)  time: 0.4992  data: 0.1460  max mem: 21002
Epoch: [266]  [130/312]  eta: 0:01:49  lr: 0.000141  min_lr: 0.000141  loss: 3.7534 (3.6045)  weight_decay: 0.0500 (0.0500)  time: 0.6626  data: 0.2727  max mem: 21002
Epoch: [266]  [140/312]  eta: 0:01:39  lr: 0.000141  min_lr: 0.000141  loss: 3.5795 (3.5924)  weight_decay: 0.0500 (0.0500)  time: 0.4483  data: 0.1281  max mem: 21002
Epoch: [266]  [150/312]  eta: 0:01:34  lr: 0.000141  min_lr: 0.000141  loss: 3.4631 (3.5781)  weight_decay: 0.0500 (0.0500)  time: 0.5039  data: 0.1433  max mem: 21002
Epoch: [266]  [160/312]  eta: 0:01:28  lr: 0.000140  min_lr: 0.000140  loss: 3.6549 (3.5900)  weight_decay: 0.0500 (0.0500)  time: 0.6185  data: 0.2190  max mem: 21002
Epoch: [266]  [170/312]  eta: 0:01:21  lr: 0.000140  min_lr: 0.000140  loss: 3.8221 (3.6109)  weight_decay: 0.0500 (0.0500)  time: 0.4848  data: 0.1145  max mem: 21002
Epoch: [266]  [180/312]  eta: 0:01:16  lr: 0.000140  min_lr: 0.000140  loss: 3.8918 (3.6137)  weight_decay: 0.0500 (0.0500)  time: 0.5684  data: 0.1436  max mem: 21002
Epoch: [266]  [190/312]  eta: 0:01:08  lr: 0.000140  min_lr: 0.000140  loss: 3.6102 (3.6023)  weight_decay: 0.0500 (0.0500)  time: 0.4847  data: 0.1056  max mem: 21002
Epoch: [266]  [200/312]  eta: 0:01:03  lr: 0.000139  min_lr: 0.000139  loss: 3.2568 (3.5896)  weight_decay: 0.0500 (0.0500)  time: 0.4735  data: 0.1426  max mem: 21002
Epoch: [266]  [210/312]  eta: 0:00:58  lr: 0.000139  min_lr: 0.000139  loss: 3.6082 (3.5956)  weight_decay: 0.0500 (0.0500)  time: 0.6303  data: 0.2999  max mem: 21002
Epoch: [266]  [220/312]  eta: 0:00:51  lr: 0.000139  min_lr: 0.000139  loss: 3.8490 (3.5965)  weight_decay: 0.0500 (0.0500)  time: 0.4442  data: 0.1581  max mem: 21002
Epoch: [266]  [230/312]  eta: 0:00:46  lr: 0.000139  min_lr: 0.000139  loss: 3.4182 (3.5813)  weight_decay: 0.0500 (0.0500)  time: 0.4788  data: 0.1629  max mem: 21002
Epoch: [266]  [240/312]  eta: 0:00:40  lr: 0.000138  min_lr: 0.000138  loss: 3.1601 (3.5733)  weight_decay: 0.0500 (0.0500)  time: 0.6520  data: 0.3364  max mem: 21002
Epoch: [266]  [250/312]  eta: 0:00:34  lr: 0.000138  min_lr: 0.000138  loss: 3.5391 (3.5665)  weight_decay: 0.0500 (0.0500)  time: 0.4594  data: 0.1742  max mem: 21002
Epoch: [266]  [260/312]  eta: 0:00:29  lr: 0.000138  min_lr: 0.000138  loss: 3.2238 (3.5492)  weight_decay: 0.0500 (0.0500)  time: 0.5126  data: 0.2120  max mem: 21002
Epoch: [266]  [270/312]  eta: 0:00:23  lr: 0.000138  min_lr: 0.000138  loss: 3.1979 (3.5438)  weight_decay: 0.0500 (0.0500)  time: 0.5124  data: 0.2122  max mem: 21002
Epoch: [266]  [280/312]  eta: 0:00:17  lr: 0.000137  min_lr: 0.000137  loss: 3.5763 (3.5459)  weight_decay: 0.0500 (0.0500)  time: 0.4626  data: 0.1560  max mem: 21002
Epoch: [266]  [290/312]  eta: 0:00:12  lr: 0.000137  min_lr: 0.000137  loss: 3.7155 (3.5439)  weight_decay: 0.0500 (0.0500)  time: 0.6330  data: 0.2836  max mem: 21002
Epoch: [266]  [300/312]  eta: 0:00:06  lr: 0.000137  min_lr: 0.000137  loss: 3.7155 (3.5448)  weight_decay: 0.0500 (0.0500)  time: 0.4520  data: 0.1283  max mem: 21002
Epoch: [266]  [310/312]  eta: 0:00:01  lr: 0.000137  min_lr: 0.000137  loss: 3.7442 (3.5473)  weight_decay: 0.0500 (0.0500)  time: 0.2775  data: 0.0001  max mem: 21002
Epoch: [266]  [311/312]  eta: 0:00:00  lr: 0.000137  min_lr: 0.000137  loss: 3.7203 (3.5476)  weight_decay: 0.0500 (0.0500)  time: 0.2774  data: 0.0001  max mem: 21002
Epoch: [266] Total time: 0:02:49 (0.5423 s / it)
Averaged stats: lr: 0.000137  min_lr: 0.000137  loss: 3.7203 (3.5841)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.1084 (1.1084)  acc1: 80.5990 (80.5990)  acc5: 95.4427 (95.4427)  time: 8.7021  data: 8.5832  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4036 (1.3195)  acc1: 73.5677 (73.7600)  acc5: 91.7969 (91.7120)  time: 1.0521  data: 0.9538  max mem: 21002
Test: Total time: 0:00:09 (1.0674 s / it)
* Acc@1 73.372 Acc@5 91.692 loss 1.327
Accuracy of the model on the 50000 test images: 73.4%
Max accuracy: 73.43%
Epoch: [267]  [  0/312]  eta: 1:02:12  lr: 0.000136  min_lr: 0.000136  loss: 3.1321 (3.1321)  weight_decay: 0.0500 (0.0500)  time: 11.9624  data: 8.8726  max mem: 21002
Epoch: [267]  [ 10/312]  eta: 0:08:50  lr: 0.000136  min_lr: 0.000136  loss: 3.8266 (3.7410)  weight_decay: 0.0500 (0.0500)  time: 1.7570  data: 1.1047  max mem: 21002
Epoch: [267]  [ 20/312]  eta: 0:05:10  lr: 0.000136  min_lr: 0.000136  loss: 3.6187 (3.5736)  weight_decay: 0.0500 (0.0500)  time: 0.5192  data: 0.1735  max mem: 21002
Epoch: [267]  [ 30/312]  eta: 0:03:50  lr: 0.000136  min_lr: 0.000136  loss: 3.8184 (3.7144)  weight_decay: 0.0500 (0.0500)  time: 0.2994  data: 0.0098  max mem: 21002
Epoch: [267]  [ 40/312]  eta: 0:03:07  lr: 0.000135  min_lr: 0.000135  loss: 3.8266 (3.6832)  weight_decay: 0.0500 (0.0500)  time: 0.2943  data: 0.0006  max mem: 21002
Epoch: [267]  [ 50/312]  eta: 0:02:54  lr: 0.000135  min_lr: 0.000135  loss: 3.7865 (3.6680)  weight_decay: 0.0500 (0.0500)  time: 0.4334  data: 0.0525  max mem: 21002
Epoch: [267]  [ 60/312]  eta: 0:02:39  lr: 0.000135  min_lr: 0.000135  loss: 3.6645 (3.6263)  weight_decay: 0.0500 (0.0500)  time: 0.5246  data: 0.1002  max mem: 21002
Epoch: [267]  [ 70/312]  eta: 0:02:30  lr: 0.000135  min_lr: 0.000135  loss: 3.6645 (3.6211)  weight_decay: 0.0500 (0.0500)  time: 0.5138  data: 0.0881  max mem: 21002
Epoch: [267]  [ 80/312]  eta: 0:02:23  lr: 0.000134  min_lr: 0.000134  loss: 3.7298 (3.6100)  weight_decay: 0.0500 (0.0500)  time: 0.5631  data: 0.1027  max mem: 21002
Epoch: [267]  [ 90/312]  eta: 0:02:13  lr: 0.000134  min_lr: 0.000134  loss: 3.7169 (3.5890)  weight_decay: 0.0500 (0.0500)  time: 0.5213  data: 0.1315  max mem: 21002
Epoch: [267]  [100/312]  eta: 0:02:06  lr: 0.000134  min_lr: 0.000134  loss: 3.2267 (3.5612)  weight_decay: 0.0500 (0.0500)  time: 0.5218  data: 0.1075  max mem: 21002
Epoch: [267]  [110/312]  eta: 0:01:59  lr: 0.000134  min_lr: 0.000134  loss: 3.2267 (3.5490)  weight_decay: 0.0500 (0.0500)  time: 0.5483  data: 0.0822  max mem: 21002
Epoch: [267]  [120/312]  eta: 0:01:51  lr: 0.000133  min_lr: 0.000133  loss: 3.5801 (3.5567)  weight_decay: 0.0500 (0.0500)  time: 0.5023  data: 0.0554  max mem: 21002
Epoch: [267]  [130/312]  eta: 0:01:46  lr: 0.000133  min_lr: 0.000133  loss: 3.8087 (3.5697)  weight_decay: 0.0500 (0.0500)  time: 0.5432  data: 0.0591  max mem: 21002
Epoch: [267]  [140/312]  eta: 0:01:40  lr: 0.000133  min_lr: 0.000133  loss: 3.8255 (3.5836)  weight_decay: 0.0500 (0.0500)  time: 0.5989  data: 0.0697  max mem: 21002
Epoch: [267]  [150/312]  eta: 0:01:33  lr: 0.000133  min_lr: 0.000133  loss: 3.9414 (3.5941)  weight_decay: 0.0500 (0.0500)  time: 0.5274  data: 0.0679  max mem: 21002
Epoch: [267]  [160/312]  eta: 0:01:27  lr: 0.000132  min_lr: 0.000132  loss: 3.6693 (3.5947)  weight_decay: 0.0500 (0.0500)  time: 0.5331  data: 0.0983  max mem: 21002
Epoch: [267]  [170/312]  eta: 0:01:21  lr: 0.000132  min_lr: 0.000132  loss: 3.6693 (3.5866)  weight_decay: 0.0500 (0.0500)  time: 0.5545  data: 0.0851  max mem: 21002
Epoch: [267]  [180/312]  eta: 0:01:15  lr: 0.000132  min_lr: 0.000132  loss: 3.8087 (3.5800)  weight_decay: 0.0500 (0.0500)  time: 0.5054  data: 0.0860  max mem: 21002
Epoch: [267]  [190/312]  eta: 0:01:09  lr: 0.000132  min_lr: 0.000132  loss: 3.4498 (3.5746)  weight_decay: 0.0500 (0.0500)  time: 0.5604  data: 0.0851  max mem: 21002
Epoch: [267]  [200/312]  eta: 0:01:03  lr: 0.000131  min_lr: 0.000131  loss: 3.5142 (3.5788)  weight_decay: 0.0500 (0.0500)  time: 0.5406  data: 0.0658  max mem: 21002
Epoch: [267]  [210/312]  eta: 0:00:57  lr: 0.000131  min_lr: 0.000131  loss: 3.4480 (3.5729)  weight_decay: 0.0500 (0.0500)  time: 0.5202  data: 0.0834  max mem: 21002
Epoch: [267]  [220/312]  eta: 0:00:52  lr: 0.000131  min_lr: 0.000131  loss: 3.6915 (3.5837)  weight_decay: 0.0500 (0.0500)  time: 0.6090  data: 0.0625  max mem: 21002
Epoch: [267]  [230/312]  eta: 0:00:46  lr: 0.000131  min_lr: 0.000131  loss: 3.7430 (3.5805)  weight_decay: 0.0500 (0.0500)  time: 0.5259  data: 0.0643  max mem: 21002
Epoch: [267]  [240/312]  eta: 0:00:40  lr: 0.000130  min_lr: 0.000130  loss: 3.5969 (3.5828)  weight_decay: 0.0500 (0.0500)  time: 0.5211  data: 0.0858  max mem: 21002
Epoch: [267]  [250/312]  eta: 0:00:34  lr: 0.000130  min_lr: 0.000130  loss: 3.7867 (3.5825)  weight_decay: 0.0500 (0.0500)  time: 0.5068  data: 0.0419  max mem: 21002
Epoch: [267]  [260/312]  eta: 0:00:29  lr: 0.000130  min_lr: 0.000130  loss: 3.6494 (3.5843)  weight_decay: 0.0500 (0.0500)  time: 0.4936  data: 0.0393  max mem: 21002
Epoch: [267]  [270/312]  eta: 0:00:23  lr: 0.000130  min_lr: 0.000130  loss: 3.7890 (3.5982)  weight_decay: 0.0500 (0.0500)  time: 0.6068  data: 0.0558  max mem: 21002
Epoch: [267]  [280/312]  eta: 0:00:17  lr: 0.000129  min_lr: 0.000129  loss: 3.9424 (3.5979)  weight_decay: 0.0500 (0.0500)  time: 0.5611  data: 0.0519  max mem: 21002
Epoch: [267]  [290/312]  eta: 0:00:12  lr: 0.000129  min_lr: 0.000129  loss: 3.8031 (3.6036)  weight_decay: 0.0500 (0.0500)  time: 0.5213  data: 0.0799  max mem: 21002
Epoch: [267]  [300/312]  eta: 0:00:06  lr: 0.000129  min_lr: 0.000129  loss: 3.4498 (3.5848)  weight_decay: 0.0500 (0.0500)  time: 0.4874  data: 0.0603  max mem: 21002
Epoch: [267]  [310/312]  eta: 0:00:01  lr: 0.000129  min_lr: 0.000129  loss: 3.2499 (3.5802)  weight_decay: 0.0500 (0.0500)  time: 0.3631  data: 0.0090  max mem: 21002
Epoch: [267]  [311/312]  eta: 0:00:00  lr: 0.000129  min_lr: 0.000129  loss: 3.2499 (3.5804)  weight_decay: 0.0500 (0.0500)  time: 0.3628  data: 0.0090  max mem: 21002
Epoch: [267] Total time: 0:02:51 (0.5504 s / it)
Averaged stats: lr: 0.000129  min_lr: 0.000129  loss: 3.2499 (3.5898)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.0077 (1.0077)  acc1: 81.2500 (81.2500)  acc5: 95.8333 (95.8333)  time: 8.7994  data: 8.6807  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3176 (1.2585)  acc1: 72.7865 (73.6800)  acc5: 91.4062 (91.9840)  time: 1.0651  data: 0.9646  max mem: 21002
Test: Total time: 0:00:10 (1.1142 s / it)
* Acc@1 73.386 Acc@5 91.914 loss 1.265
Accuracy of the model on the 50000 test images: 73.4%
Max accuracy: 73.43%
Epoch: [268]  [  0/312]  eta: 1:00:50  lr: 0.000128  min_lr: 0.000128  loss: 3.8452 (3.8452)  weight_decay: 0.0500 (0.0500)  time: 11.6994  data: 10.7288  max mem: 21002
Epoch: [268]  [ 10/312]  eta: 0:07:39  lr: 0.000128  min_lr: 0.000128  loss: 3.5264 (3.2814)  weight_decay: 0.0500 (0.0500)  time: 1.5208  data: 1.0727  max mem: 21002
Epoch: [268]  [ 20/312]  eta: 0:05:00  lr: 0.000128  min_lr: 0.000128  loss: 3.5706 (3.4466)  weight_decay: 0.0500 (0.0500)  time: 0.4966  data: 0.1568  max mem: 21002
Epoch: [268]  [ 30/312]  eta: 0:03:43  lr: 0.000128  min_lr: 0.000128  loss: 3.6286 (3.4919)  weight_decay: 0.0500 (0.0500)  time: 0.3898  data: 0.1036  max mem: 21002
Epoch: [268]  [ 40/312]  eta: 0:03:17  lr: 0.000127  min_lr: 0.000127  loss: 3.6467 (3.4902)  weight_decay: 0.0500 (0.0500)  time: 0.4065  data: 0.1197  max mem: 21002
Epoch: [268]  [ 50/312]  eta: 0:03:08  lr: 0.000127  min_lr: 0.000127  loss: 3.6467 (3.5150)  weight_decay: 0.0500 (0.0500)  time: 0.6114  data: 0.2881  max mem: 21002
Epoch: [268]  [ 60/312]  eta: 0:02:43  lr: 0.000127  min_lr: 0.000127  loss: 3.4519 (3.4829)  weight_decay: 0.0500 (0.0500)  time: 0.4932  data: 0.1690  max mem: 21002
Epoch: [268]  [ 70/312]  eta: 0:02:36  lr: 0.000127  min_lr: 0.000127  loss: 3.5082 (3.5091)  weight_decay: 0.0500 (0.0500)  time: 0.4621  data: 0.1539  max mem: 21002
Epoch: [268]  [ 80/312]  eta: 0:02:30  lr: 0.000126  min_lr: 0.000126  loss: 3.7592 (3.5053)  weight_decay: 0.0500 (0.0500)  time: 0.6511  data: 0.3348  max mem: 21002
Epoch: [268]  [ 90/312]  eta: 0:02:17  lr: 0.000126  min_lr: 0.000126  loss: 3.4259 (3.4864)  weight_decay: 0.0500 (0.0500)  time: 0.5105  data: 0.2156  max mem: 21002
Epoch: [268]  [100/312]  eta: 0:02:12  lr: 0.000126  min_lr: 0.000126  loss: 3.5081 (3.4773)  weight_decay: 0.0500 (0.0500)  time: 0.5285  data: 0.1983  max mem: 21002
Epoch: [268]  [110/312]  eta: 0:02:00  lr: 0.000126  min_lr: 0.000126  loss: 3.5755 (3.4855)  weight_decay: 0.0500 (0.0500)  time: 0.4954  data: 0.1641  max mem: 21002
Epoch: [268]  [120/312]  eta: 0:01:55  lr: 0.000125  min_lr: 0.000125  loss: 3.6802 (3.4811)  weight_decay: 0.0500 (0.0500)  time: 0.4910  data: 0.1236  max mem: 21002
Epoch: [268]  [130/312]  eta: 0:01:50  lr: 0.000125  min_lr: 0.000125  loss: 3.6802 (3.5027)  weight_decay: 0.0500 (0.0500)  time: 0.6669  data: 0.2559  max mem: 21002
Epoch: [268]  [140/312]  eta: 0:01:40  lr: 0.000125  min_lr: 0.000125  loss: 3.8275 (3.5106)  weight_decay: 0.0500 (0.0500)  time: 0.4635  data: 0.1328  max mem: 21002
Epoch: [268]  [150/312]  eta: 0:01:36  lr: 0.000125  min_lr: 0.000125  loss: 3.5602 (3.5030)  weight_decay: 0.0500 (0.0500)  time: 0.5307  data: 0.1693  max mem: 21002
Epoch: [268]  [160/312]  eta: 0:01:29  lr: 0.000124  min_lr: 0.000124  loss: 3.4545 (3.4974)  weight_decay: 0.0500 (0.0500)  time: 0.6388  data: 0.2363  max mem: 21002
Epoch: [268]  [170/312]  eta: 0:01:21  lr: 0.000124  min_lr: 0.000124  loss: 3.4545 (3.4924)  weight_decay: 0.0500 (0.0500)  time: 0.4264  data: 0.0964  max mem: 21002
Epoch: [268]  [180/312]  eta: 0:01:16  lr: 0.000124  min_lr: 0.000124  loss: 3.7358 (3.4983)  weight_decay: 0.0500 (0.0500)  time: 0.5058  data: 0.1815  max mem: 21002
Epoch: [268]  [190/312]  eta: 0:01:09  lr: 0.000124  min_lr: 0.000124  loss: 3.7427 (3.5124)  weight_decay: 0.0500 (0.0500)  time: 0.4732  data: 0.1527  max mem: 21002
Epoch: [268]  [200/312]  eta: 0:01:04  lr: 0.000123  min_lr: 0.000123  loss: 3.7180 (3.5162)  weight_decay: 0.0500 (0.0500)  time: 0.5025  data: 0.1831  max mem: 21002
Epoch: [268]  [210/312]  eta: 0:00:59  lr: 0.000123  min_lr: 0.000123  loss: 3.5351 (3.5135)  weight_decay: 0.0500 (0.0500)  time: 0.7067  data: 0.3250  max mem: 21002
Epoch: [268]  [220/312]  eta: 0:00:52  lr: 0.000123  min_lr: 0.000123  loss: 3.4878 (3.5122)  weight_decay: 0.0500 (0.0500)  time: 0.4894  data: 0.1429  max mem: 21002
Epoch: [268]  [230/312]  eta: 0:00:46  lr: 0.000123  min_lr: 0.000123  loss: 3.7260 (3.5147)  weight_decay: 0.0500 (0.0500)  time: 0.4896  data: 0.1236  max mem: 21002
Epoch: [268]  [240/312]  eta: 0:00:41  lr: 0.000123  min_lr: 0.000123  loss: 3.6443 (3.5176)  weight_decay: 0.0500 (0.0500)  time: 0.6360  data: 0.1799  max mem: 21002
Epoch: [268]  [250/312]  eta: 0:00:35  lr: 0.000122  min_lr: 0.000122  loss: 3.3092 (3.5136)  weight_decay: 0.0500 (0.0500)  time: 0.5073  data: 0.1337  max mem: 21002
Epoch: [268]  [260/312]  eta: 0:00:29  lr: 0.000122  min_lr: 0.000122  loss: 3.3153 (3.5133)  weight_decay: 0.0500 (0.0500)  time: 0.5550  data: 0.1977  max mem: 21002
Epoch: [268]  [270/312]  eta: 0:00:23  lr: 0.000122  min_lr: 0.000122  loss: 3.6701 (3.5122)  weight_decay: 0.0500 (0.0500)  time: 0.4799  data: 0.1214  max mem: 21002
Epoch: [268]  [280/312]  eta: 0:00:18  lr: 0.000122  min_lr: 0.000122  loss: 3.8565 (3.5242)  weight_decay: 0.0500 (0.0500)  time: 0.4878  data: 0.1084  max mem: 21002
Epoch: [268]  [290/312]  eta: 0:00:12  lr: 0.000121  min_lr: 0.000121  loss: 3.8565 (3.5248)  weight_decay: 0.0500 (0.0500)  time: 0.6034  data: 0.1666  max mem: 21002
Epoch: [268]  [300/312]  eta: 0:00:06  lr: 0.000121  min_lr: 0.000121  loss: 3.6081 (3.5250)  weight_decay: 0.0500 (0.0500)  time: 0.3983  data: 0.0587  max mem: 21002
Epoch: [268]  [310/312]  eta: 0:00:01  lr: 0.000121  min_lr: 0.000121  loss: 3.4810 (3.5199)  weight_decay: 0.0500 (0.0500)  time: 0.2776  data: 0.0001  max mem: 21002
Epoch: [268]  [311/312]  eta: 0:00:00  lr: 0.000121  min_lr: 0.000121  loss: 3.5202 (3.5207)  weight_decay: 0.0500 (0.0500)  time: 0.2775  data: 0.0001  max mem: 21002
Epoch: [268] Total time: 0:02:50 (0.5473 s / it)
Averaged stats: lr: 0.000121  min_lr: 0.000121  loss: 3.5202 (3.5960)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.0071 (1.0071)  acc1: 82.1615 (82.1615)  acc5: 95.5729 (95.5729)  time: 8.8362  data: 8.7176  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3402 (1.2585)  acc1: 72.2656 (73.8560)  acc5: 91.4062 (91.6320)  time: 1.0673  data: 0.9687  max mem: 21002
Test: Total time: 0:00:09 (1.0783 s / it)
* Acc@1 73.342 Acc@5 91.832 loss 1.267
Accuracy of the model on the 50000 test images: 73.3%
Max accuracy: 73.43%
Epoch: [269]  [  0/312]  eta: 1:01:40  lr: 0.000121  min_lr: 0.000121  loss: 3.3006 (3.3006)  weight_decay: 0.0500 (0.0500)  time: 11.8612  data: 10.9832  max mem: 21002
Epoch: [269]  [ 10/312]  eta: 0:08:03  lr: 0.000120  min_lr: 0.000120  loss: 3.8350 (3.6855)  weight_decay: 0.0500 (0.0500)  time: 1.6009  data: 0.9990  max mem: 21002
Epoch: [269]  [ 20/312]  eta: 0:05:08  lr: 0.000120  min_lr: 0.000120  loss: 3.7593 (3.6140)  weight_decay: 0.0500 (0.0500)  time: 0.5146  data: 0.0868  max mem: 21002
Epoch: [269]  [ 30/312]  eta: 0:03:48  lr: 0.000120  min_lr: 0.000120  loss: 3.6616 (3.5713)  weight_decay: 0.0500 (0.0500)  time: 0.3737  data: 0.0868  max mem: 21002
Epoch: [269]  [ 40/312]  eta: 0:03:16  lr: 0.000120  min_lr: 0.000120  loss: 3.6616 (3.5946)  weight_decay: 0.0500 (0.0500)  time: 0.3744  data: 0.0778  max mem: 21002
Epoch: [269]  [ 50/312]  eta: 0:03:04  lr: 0.000120  min_lr: 0.000120  loss: 3.8604 (3.6519)  weight_decay: 0.0500 (0.0500)  time: 0.5448  data: 0.2287  max mem: 21002
Epoch: [269]  [ 60/312]  eta: 0:02:42  lr: 0.000119  min_lr: 0.000119  loss: 3.8604 (3.6593)  weight_decay: 0.0500 (0.0500)  time: 0.4845  data: 0.1515  max mem: 21002
Epoch: [269]  [ 70/312]  eta: 0:02:37  lr: 0.000119  min_lr: 0.000119  loss: 3.7876 (3.6696)  weight_decay: 0.0500 (0.0500)  time: 0.5085  data: 0.1704  max mem: 21002
Epoch: [269]  [ 80/312]  eta: 0:02:29  lr: 0.000119  min_lr: 0.000119  loss: 3.8730 (3.6975)  weight_decay: 0.0500 (0.0500)  time: 0.6443  data: 0.3322  max mem: 21002
Epoch: [269]  [ 90/312]  eta: 0:02:16  lr: 0.000119  min_lr: 0.000119  loss: 3.7051 (3.6713)  weight_decay: 0.0500 (0.0500)  time: 0.4890  data: 0.1624  max mem: 21002
Epoch: [269]  [100/312]  eta: 0:02:11  lr: 0.000118  min_lr: 0.000118  loss: 3.7003 (3.6754)  weight_decay: 0.0500 (0.0500)  time: 0.5264  data: 0.1507  max mem: 21002
Epoch: [269]  [110/312]  eta: 0:01:59  lr: 0.000118  min_lr: 0.000118  loss: 3.5530 (3.6520)  weight_decay: 0.0500 (0.0500)  time: 0.4826  data: 0.1506  max mem: 21002
Epoch: [269]  [120/312]  eta: 0:01:54  lr: 0.000118  min_lr: 0.000118  loss: 3.7181 (3.6611)  weight_decay: 0.0500 (0.0500)  time: 0.4776  data: 0.1501  max mem: 21002
Epoch: [269]  [130/312]  eta: 0:01:49  lr: 0.000118  min_lr: 0.000118  loss: 3.7905 (3.6580)  weight_decay: 0.0500 (0.0500)  time: 0.6466  data: 0.2923  max mem: 21002
Epoch: [269]  [140/312]  eta: 0:01:40  lr: 0.000117  min_lr: 0.000117  loss: 3.7340 (3.6690)  weight_decay: 0.0500 (0.0500)  time: 0.5051  data: 0.1429  max mem: 21002
Epoch: [269]  [150/312]  eta: 0:01:35  lr: 0.000117  min_lr: 0.000117  loss: 3.6788 (3.6549)  weight_decay: 0.0500 (0.0500)  time: 0.5389  data: 0.1430  max mem: 21002
Epoch: [269]  [160/312]  eta: 0:01:29  lr: 0.000117  min_lr: 0.000117  loss: 3.6735 (3.6546)  weight_decay: 0.0500 (0.0500)  time: 0.5997  data: 0.2456  max mem: 21002
Epoch: [269]  [170/312]  eta: 0:01:22  lr: 0.000117  min_lr: 0.000117  loss: 3.8736 (3.6589)  weight_decay: 0.0500 (0.0500)  time: 0.4779  data: 0.1035  max mem: 21002
Epoch: [269]  [180/312]  eta: 0:01:16  lr: 0.000116  min_lr: 0.000116  loss: 3.8227 (3.6578)  weight_decay: 0.0500 (0.0500)  time: 0.5537  data: 0.1069  max mem: 21002
Epoch: [269]  [190/312]  eta: 0:01:09  lr: 0.000116  min_lr: 0.000116  loss: 3.5059 (3.6554)  weight_decay: 0.0500 (0.0500)  time: 0.4751  data: 0.1076  max mem: 21002
Epoch: [269]  [200/312]  eta: 0:01:04  lr: 0.000116  min_lr: 0.000116  loss: 3.5059 (3.6471)  weight_decay: 0.0500 (0.0500)  time: 0.4924  data: 0.0930  max mem: 21002
Epoch: [269]  [210/312]  eta: 0:00:58  lr: 0.000116  min_lr: 0.000116  loss: 3.8394 (3.6526)  weight_decay: 0.0500 (0.0500)  time: 0.6724  data: 0.1865  max mem: 21002
Epoch: [269]  [220/312]  eta: 0:00:52  lr: 0.000115  min_lr: 0.000115  loss: 3.8333 (3.6462)  weight_decay: 0.0500 (0.0500)  time: 0.4822  data: 0.0957  max mem: 21002
Epoch: [269]  [230/312]  eta: 0:00:46  lr: 0.000115  min_lr: 0.000115  loss: 3.8213 (3.6507)  weight_decay: 0.0500 (0.0500)  time: 0.5177  data: 0.0766  max mem: 21002
Epoch: [269]  [240/312]  eta: 0:00:40  lr: 0.000115  min_lr: 0.000115  loss: 3.6893 (3.6472)  weight_decay: 0.0500 (0.0500)  time: 0.6007  data: 0.1219  max mem: 21002
Epoch: [269]  [250/312]  eta: 0:00:35  lr: 0.000115  min_lr: 0.000115  loss: 3.6893 (3.6502)  weight_decay: 0.0500 (0.0500)  time: 0.5048  data: 0.0467  max mem: 21002
Epoch: [269]  [260/312]  eta: 0:00:29  lr: 0.000114  min_lr: 0.000114  loss: 3.8436 (3.6525)  weight_decay: 0.0500 (0.0500)  time: 0.6058  data: 0.0402  max mem: 21002
Epoch: [269]  [270/312]  eta: 0:00:23  lr: 0.000114  min_lr: 0.000114  loss: 3.5132 (3.6358)  weight_decay: 0.0500 (0.0500)  time: 0.4893  data: 0.0402  max mem: 21002
Epoch: [269]  [280/312]  eta: 0:00:18  lr: 0.000114  min_lr: 0.000114  loss: 3.3109 (3.6332)  weight_decay: 0.0500 (0.0500)  time: 0.4926  data: 0.0408  max mem: 21002
Epoch: [269]  [290/312]  eta: 0:00:12  lr: 0.000114  min_lr: 0.000114  loss: 3.5631 (3.6288)  weight_decay: 0.0500 (0.0500)  time: 0.5627  data: 0.0445  max mem: 21002
Epoch: [269]  [300/312]  eta: 0:00:06  lr: 0.000113  min_lr: 0.000113  loss: 3.7309 (3.6300)  weight_decay: 0.0500 (0.0500)  time: 0.3813  data: 0.0042  max mem: 21002
Epoch: [269]  [310/312]  eta: 0:00:01  lr: 0.000113  min_lr: 0.000113  loss: 3.8408 (3.6257)  weight_decay: 0.0500 (0.0500)  time: 0.3022  data: 0.0001  max mem: 21002
Epoch: [269]  [311/312]  eta: 0:00:00  lr: 0.000113  min_lr: 0.000113  loss: 3.8408 (3.6276)  weight_decay: 0.0500 (0.0500)  time: 0.2772  data: 0.0001  max mem: 21002
Epoch: [269] Total time: 0:02:50 (0.5474 s / it)
Averaged stats: lr: 0.000113  min_lr: 0.000113  loss: 3.8408 (3.5778)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.0543 (1.0543)  acc1: 81.1198 (81.1198)  acc5: 95.7031 (95.7031)  time: 8.4674  data: 8.3491  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4124 (1.3008)  acc1: 71.8750 (73.3120)  acc5: 91.1458 (91.6960)  time: 1.0241  data: 0.9278  max mem: 21002
Test: Total time: 0:00:09 (1.0361 s / it)
* Acc@1 73.454 Acc@5 91.890 loss 1.308
Accuracy of the model on the 50000 test images: 73.5%
Max accuracy: 73.45%
Epoch: [270]  [  0/312]  eta: 1:00:46  lr: 0.000113  min_lr: 0.000113  loss: 3.3902 (3.3902)  weight_decay: 0.0500 (0.0500)  time: 11.6865  data: 11.3696  max mem: 21002
Epoch: [270]  [ 10/312]  eta: 0:07:59  lr: 0.000113  min_lr: 0.000113  loss: 3.6925 (3.6341)  weight_decay: 0.0500 (0.0500)  time: 1.5866  data: 1.0592  max mem: 21002
Epoch: [270]  [ 20/312]  eta: 0:05:10  lr: 0.000113  min_lr: 0.000113  loss: 3.7592 (3.7540)  weight_decay: 0.0500 (0.0500)  time: 0.5305  data: 0.0774  max mem: 21002
Epoch: [270]  [ 30/312]  eta: 0:03:49  lr: 0.000112  min_lr: 0.000112  loss: 3.8670 (3.7462)  weight_decay: 0.0500 (0.0500)  time: 0.3897  data: 0.0645  max mem: 21002
Epoch: [270]  [ 40/312]  eta: 0:03:15  lr: 0.000112  min_lr: 0.000112  loss: 3.7367 (3.7123)  weight_decay: 0.0500 (0.0500)  time: 0.3565  data: 0.0663  max mem: 21002
Epoch: [270]  [ 50/312]  eta: 0:03:10  lr: 0.000112  min_lr: 0.000112  loss: 3.8717 (3.7207)  weight_decay: 0.0500 (0.0500)  time: 0.5931  data: 0.2544  max mem: 21002
Epoch: [270]  [ 60/312]  eta: 0:02:45  lr: 0.000112  min_lr: 0.000112  loss: 3.6916 (3.6950)  weight_decay: 0.0500 (0.0500)  time: 0.5267  data: 0.1896  max mem: 21002
Epoch: [270]  [ 70/312]  eta: 0:02:37  lr: 0.000112  min_lr: 0.000112  loss: 3.5916 (3.6818)  weight_decay: 0.0500 (0.0500)  time: 0.4525  data: 0.1026  max mem: 21002
Epoch: [270]  [ 80/312]  eta: 0:02:26  lr: 0.000111  min_lr: 0.000111  loss: 3.7876 (3.6601)  weight_decay: 0.0500 (0.0500)  time: 0.5556  data: 0.1869  max mem: 21002
Epoch: [270]  [ 90/312]  eta: 0:02:16  lr: 0.000111  min_lr: 0.000111  loss: 3.6491 (3.6358)  weight_decay: 0.0500 (0.0500)  time: 0.4906  data: 0.1228  max mem: 21002
Epoch: [270]  [100/312]  eta: 0:02:10  lr: 0.000111  min_lr: 0.000111  loss: 3.4453 (3.6193)  weight_decay: 0.0500 (0.0500)  time: 0.5542  data: 0.1090  max mem: 21002
Epoch: [270]  [110/312]  eta: 0:01:58  lr: 0.000111  min_lr: 0.000111  loss: 3.4453 (3.6019)  weight_decay: 0.0500 (0.0500)  time: 0.4578  data: 0.0742  max mem: 21002
Epoch: [270]  [120/312]  eta: 0:01:54  lr: 0.000110  min_lr: 0.000110  loss: 3.2366 (3.5676)  weight_decay: 0.0500 (0.0500)  time: 0.5026  data: 0.1292  max mem: 21002
Epoch: [270]  [130/312]  eta: 0:01:50  lr: 0.000110  min_lr: 0.000110  loss: 3.4559 (3.5766)  weight_decay: 0.0500 (0.0500)  time: 0.7210  data: 0.2347  max mem: 21002
Epoch: [270]  [140/312]  eta: 0:01:40  lr: 0.000110  min_lr: 0.000110  loss: 3.8810 (3.6048)  weight_decay: 0.0500 (0.0500)  time: 0.5103  data: 0.1092  max mem: 21002
Epoch: [270]  [150/312]  eta: 0:01:35  lr: 0.000110  min_lr: 0.000110  loss: 3.9248 (3.6144)  weight_decay: 0.0500 (0.0500)  time: 0.4610  data: 0.0508  max mem: 21002
Epoch: [270]  [160/312]  eta: 0:01:29  lr: 0.000109  min_lr: 0.000109  loss: 3.7759 (3.6254)  weight_decay: 0.0500 (0.0500)  time: 0.6023  data: 0.0516  max mem: 21002
Epoch: [270]  [170/312]  eta: 0:01:21  lr: 0.000109  min_lr: 0.000109  loss: 3.5980 (3.6241)  weight_decay: 0.0500 (0.0500)  time: 0.4909  data: 0.0508  max mem: 21002
Epoch: [270]  [180/312]  eta: 0:01:17  lr: 0.000109  min_lr: 0.000109  loss: 3.4990 (3.6276)  weight_decay: 0.0500 (0.0500)  time: 0.5638  data: 0.1125  max mem: 21002
Epoch: [270]  [190/312]  eta: 0:01:09  lr: 0.000109  min_lr: 0.000109  loss: 3.7611 (3.6300)  weight_decay: 0.0500 (0.0500)  time: 0.5011  data: 0.0635  max mem: 21002
Epoch: [270]  [200/312]  eta: 0:01:04  lr: 0.000109  min_lr: 0.000109  loss: 3.6610 (3.6310)  weight_decay: 0.0500 (0.0500)  time: 0.4835  data: 0.0530  max mem: 21002
Epoch: [270]  [210/312]  eta: 0:00:59  lr: 0.000108  min_lr: 0.000108  loss: 3.6610 (3.6348)  weight_decay: 0.0500 (0.0500)  time: 0.6774  data: 0.0989  max mem: 21002
Epoch: [270]  [220/312]  eta: 0:00:52  lr: 0.000108  min_lr: 0.000108  loss: 3.5337 (3.6295)  weight_decay: 0.0500 (0.0500)  time: 0.4846  data: 0.0515  max mem: 21002
Epoch: [270]  [230/312]  eta: 0:00:47  lr: 0.000108  min_lr: 0.000108  loss: 3.4073 (3.6259)  weight_decay: 0.0500 (0.0500)  time: 0.5142  data: 0.0653  max mem: 21002
Epoch: [270]  [240/312]  eta: 0:00:41  lr: 0.000108  min_lr: 0.000108  loss: 3.4864 (3.6128)  weight_decay: 0.0500 (0.0500)  time: 0.6430  data: 0.0614  max mem: 21002
Epoch: [270]  [250/312]  eta: 0:00:34  lr: 0.000107  min_lr: 0.000107  loss: 3.4562 (3.6131)  weight_decay: 0.0500 (0.0500)  time: 0.4417  data: 0.0055  max mem: 21002
Epoch: [270]  [260/312]  eta: 0:00:29  lr: 0.000107  min_lr: 0.000107  loss: 3.7413 (3.6146)  weight_decay: 0.0500 (0.0500)  time: 0.5075  data: 0.0437  max mem: 21002
Epoch: [270]  [270/312]  eta: 0:00:23  lr: 0.000107  min_lr: 0.000107  loss: 3.7832 (3.6209)  weight_decay: 0.0500 (0.0500)  time: 0.4966  data: 0.0397  max mem: 21002
Epoch: [270]  [280/312]  eta: 0:00:18  lr: 0.000107  min_lr: 0.000107  loss: 3.7431 (3.6182)  weight_decay: 0.0500 (0.0500)  time: 0.5143  data: 0.0093  max mem: 21002
Epoch: [270]  [290/312]  eta: 0:00:12  lr: 0.000106  min_lr: 0.000106  loss: 3.6331 (3.6229)  weight_decay: 0.0500 (0.0500)  time: 0.6247  data: 0.0199  max mem: 21002
Epoch: [270]  [300/312]  eta: 0:00:06  lr: 0.000106  min_lr: 0.000106  loss: 3.8603 (3.6236)  weight_decay: 0.0500 (0.0500)  time: 0.4171  data: 0.0248  max mem: 21002
Epoch: [270]  [310/312]  eta: 0:00:01  lr: 0.000106  min_lr: 0.000106  loss: 3.7463 (3.6268)  weight_decay: 0.0500 (0.0500)  time: 0.2943  data: 0.0138  max mem: 21002
Epoch: [270]  [311/312]  eta: 0:00:00  lr: 0.000106  min_lr: 0.000106  loss: 3.7335 (3.6230)  weight_decay: 0.0500 (0.0500)  time: 0.2942  data: 0.0138  max mem: 21002
Epoch: [270] Total time: 0:02:51 (0.5482 s / it)
Averaged stats: lr: 0.000106  min_lr: 0.000106  loss: 3.7335 (3.5974)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.0351 (1.0351)  acc1: 81.2500 (81.2500)  acc5: 95.4427 (95.4427)  time: 8.8071  data: 8.6898  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3834 (1.3025)  acc1: 72.1354 (73.4720)  acc5: 91.2760 (91.7120)  time: 1.0671  data: 0.9656  max mem: 21002
Test: Total time: 0:00:09 (1.0897 s / it)
* Acc@1 73.200 Acc@5 91.874 loss 1.307
Accuracy of the model on the 50000 test images: 73.2%
Max accuracy: 73.45%
Epoch: [271]  [  0/312]  eta: 0:59:07  lr: 0.000106  min_lr: 0.000106  loss: 2.3529 (2.3529)  weight_decay: 0.0500 (0.0500)  time: 11.3692  data: 8.4362  max mem: 21002
Epoch: [271]  [ 10/312]  eta: 0:07:57  lr: 0.000106  min_lr: 0.000106  loss: 3.3594 (3.3035)  weight_decay: 0.0500 (0.0500)  time: 1.5816  data: 1.0043  max mem: 21002
Epoch: [271]  [ 20/312]  eta: 0:05:28  lr: 0.000105  min_lr: 0.000105  loss: 3.6362 (3.5156)  weight_decay: 0.0500 (0.0500)  time: 0.6135  data: 0.1776  max mem: 21002
Epoch: [271]  [ 30/312]  eta: 0:04:01  lr: 0.000105  min_lr: 0.000105  loss: 3.8549 (3.6202)  weight_decay: 0.0500 (0.0500)  time: 0.4589  data: 0.0474  max mem: 21002
Epoch: [271]  [ 40/312]  eta: 0:03:26  lr: 0.000105  min_lr: 0.000105  loss: 3.9185 (3.6483)  weight_decay: 0.0500 (0.0500)  time: 0.3765  data: 0.0008  max mem: 21002
Epoch: [271]  [ 50/312]  eta: 0:03:15  lr: 0.000105  min_lr: 0.000105  loss: 3.7389 (3.6562)  weight_decay: 0.0500 (0.0500)  time: 0.5751  data: 0.0075  max mem: 21002
Epoch: [271]  [ 60/312]  eta: 0:02:49  lr: 0.000105  min_lr: 0.000105  loss: 3.6885 (3.6114)  weight_decay: 0.0500 (0.0500)  time: 0.4985  data: 0.0187  max mem: 21002
Epoch: [271]  [ 70/312]  eta: 0:02:46  lr: 0.000104  min_lr: 0.000104  loss: 3.6448 (3.6387)  weight_decay: 0.0500 (0.0500)  time: 0.5331  data: 0.0482  max mem: 21002
Epoch: [271]  [ 80/312]  eta: 0:02:34  lr: 0.000104  min_lr: 0.000104  loss: 3.6692 (3.6308)  weight_decay: 0.0500 (0.0500)  time: 0.6442  data: 0.0433  max mem: 21002
Epoch: [271]  [ 90/312]  eta: 0:02:22  lr: 0.000104  min_lr: 0.000104  loss: 3.7157 (3.6340)  weight_decay: 0.0500 (0.0500)  time: 0.4928  data: 0.0429  max mem: 21002
Epoch: [271]  [100/312]  eta: 0:02:16  lr: 0.000104  min_lr: 0.000104  loss: 3.7258 (3.6141)  weight_decay: 0.0500 (0.0500)  time: 0.5435  data: 0.0525  max mem: 21002
Epoch: [271]  [110/312]  eta: 0:02:05  lr: 0.000103  min_lr: 0.000103  loss: 3.5534 (3.6105)  weight_decay: 0.0500 (0.0500)  time: 0.5096  data: 0.0167  max mem: 21002
Epoch: [271]  [120/312]  eta: 0:01:59  lr: 0.000103  min_lr: 0.000103  loss: 3.8534 (3.6230)  weight_decay: 0.0500 (0.0500)  time: 0.5185  data: 0.0318  max mem: 21002
Epoch: [271]  [130/312]  eta: 0:01:52  lr: 0.000103  min_lr: 0.000103  loss: 3.8922 (3.6297)  weight_decay: 0.0500 (0.0500)  time: 0.6225  data: 0.0366  max mem: 21002
Epoch: [271]  [140/312]  eta: 0:01:44  lr: 0.000103  min_lr: 0.000103  loss: 3.6690 (3.6247)  weight_decay: 0.0500 (0.0500)  time: 0.5006  data: 0.0639  max mem: 21002
Epoch: [271]  [150/312]  eta: 0:01:38  lr: 0.000102  min_lr: 0.000102  loss: 3.5418 (3.6120)  weight_decay: 0.0500 (0.0500)  time: 0.5126  data: 0.0765  max mem: 21002
Epoch: [271]  [160/312]  eta: 0:01:31  lr: 0.000102  min_lr: 0.000102  loss: 3.5418 (3.6136)  weight_decay: 0.0500 (0.0500)  time: 0.5910  data: 0.0181  max mem: 21002
Epoch: [271]  [170/312]  eta: 0:01:24  lr: 0.000102  min_lr: 0.000102  loss: 3.6361 (3.6173)  weight_decay: 0.0500 (0.0500)  time: 0.5293  data: 0.0773  max mem: 21002
Epoch: [271]  [180/312]  eta: 0:01:17  lr: 0.000102  min_lr: 0.000102  loss: 3.4565 (3.6026)  weight_decay: 0.0500 (0.0500)  time: 0.4904  data: 0.0941  max mem: 21002
Epoch: [271]  [190/312]  eta: 0:01:10  lr: 0.000102  min_lr: 0.000102  loss: 3.3657 (3.5959)  weight_decay: 0.0500 (0.0500)  time: 0.3886  data: 0.0179  max mem: 21002
Epoch: [271]  [200/312]  eta: 0:01:05  lr: 0.000101  min_lr: 0.000101  loss: 3.6952 (3.6013)  weight_decay: 0.0500 (0.0500)  time: 0.4978  data: 0.0186  max mem: 21002
Epoch: [271]  [210/312]  eta: 0:00:59  lr: 0.000101  min_lr: 0.000101  loss: 3.8338 (3.6007)  weight_decay: 0.0500 (0.0500)  time: 0.6683  data: 0.0229  max mem: 21002
Epoch: [271]  [220/312]  eta: 0:00:52  lr: 0.000101  min_lr: 0.000101  loss: 3.8338 (3.6115)  weight_decay: 0.0500 (0.0500)  time: 0.4904  data: 0.0220  max mem: 21002
Epoch: [271]  [230/312]  eta: 0:00:47  lr: 0.000101  min_lr: 0.000101  loss: 3.6898 (3.6053)  weight_decay: 0.0500 (0.0500)  time: 0.5227  data: 0.0645  max mem: 21002
Epoch: [271]  [240/312]  eta: 0:00:41  lr: 0.000100  min_lr: 0.000100  loss: 3.3987 (3.5982)  weight_decay: 0.0500 (0.0500)  time: 0.6440  data: 0.0483  max mem: 21002
Epoch: [271]  [250/312]  eta: 0:00:35  lr: 0.000100  min_lr: 0.000100  loss: 3.7428 (3.6017)  weight_decay: 0.0500 (0.0500)  time: 0.4855  data: 0.0492  max mem: 21002
Epoch: [271]  [260/312]  eta: 0:00:29  lr: 0.000100  min_lr: 0.000100  loss: 3.8300 (3.6167)  weight_decay: 0.0500 (0.0500)  time: 0.5220  data: 0.0936  max mem: 21002
Epoch: [271]  [270/312]  eta: 0:00:23  lr: 0.000100  min_lr: 0.000100  loss: 3.7919 (3.6153)  weight_decay: 0.0500 (0.0500)  time: 0.4789  data: 0.0454  max mem: 21002
Epoch: [271]  [280/312]  eta: 0:00:18  lr: 0.000100  min_lr: 0.000100  loss: 3.7471 (3.6232)  weight_decay: 0.0500 (0.0500)  time: 0.4913  data: 0.0476  max mem: 21002
Epoch: [271]  [290/312]  eta: 0:00:12  lr: 0.000099  min_lr: 0.000099  loss: 3.8128 (3.6281)  weight_decay: 0.0500 (0.0500)  time: 0.5902  data: 0.0683  max mem: 21002
Epoch: [271]  [300/312]  eta: 0:00:06  lr: 0.000099  min_lr: 0.000099  loss: 3.7565 (3.6248)  weight_decay: 0.0500 (0.0500)  time: 0.3864  data: 0.0211  max mem: 21002
Epoch: [271]  [310/312]  eta: 0:00:01  lr: 0.000099  min_lr: 0.000099  loss: 3.7379 (3.6250)  weight_decay: 0.0500 (0.0500)  time: 0.2777  data: 0.0001  max mem: 21002
Epoch: [271]  [311/312]  eta: 0:00:00  lr: 0.000099  min_lr: 0.000099  loss: 3.7379 (3.6211)  weight_decay: 0.0500 (0.0500)  time: 0.2775  data: 0.0001  max mem: 21002
Epoch: [271] Total time: 0:02:51 (0.5506 s / it)
Averaged stats: lr: 0.000099  min_lr: 0.000099  loss: 3.7379 (3.5714)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 1.0202 (1.0202)  acc1: 81.1198 (81.1198)  acc5: 95.4427 (95.4427)  time: 8.1536  data: 8.0362  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3671 (1.2670)  acc1: 73.4375 (73.9360)  acc5: 91.2760 (91.8720)  time: 1.0404  data: 0.9419  max mem: 21002
Test: Total time: 0:00:09 (1.0542 s / it)
* Acc@1 73.620 Acc@5 92.002 loss 1.274
Accuracy of the model on the 50000 test images: 73.6%
Max accuracy: 73.62%
Epoch: [272]  [  0/312]  eta: 0:57:40  lr: 0.000099  min_lr: 0.000099  loss: 3.3520 (3.3520)  weight_decay: 0.0500 (0.0500)  time: 11.0903  data: 10.2575  max mem: 21002
Epoch: [272]  [ 10/312]  eta: 0:07:00  lr: 0.000099  min_lr: 0.000099  loss: 3.8114 (3.8552)  weight_decay: 0.0500 (0.0500)  time: 1.3927  data: 1.0477  max mem: 21002
Epoch: [272]  [ 20/312]  eta: 0:05:02  lr: 0.000098  min_lr: 0.000098  loss: 3.8720 (3.8017)  weight_decay: 0.0500 (0.0500)  time: 0.5326  data: 0.2319  max mem: 21002
Epoch: [272]  [ 30/312]  eta: 0:03:44  lr: 0.000098  min_lr: 0.000098  loss: 3.5477 (3.7020)  weight_decay: 0.0500 (0.0500)  time: 0.4675  data: 0.1689  max mem: 21002
Epoch: [272]  [ 40/312]  eta: 0:03:26  lr: 0.000098  min_lr: 0.000098  loss: 3.4832 (3.6559)  weight_decay: 0.0500 (0.0500)  time: 0.4661  data: 0.1784  max mem: 21002
Epoch: [272]  [ 50/312]  eta: 0:03:15  lr: 0.000098  min_lr: 0.000098  loss: 3.6691 (3.6598)  weight_decay: 0.0500 (0.0500)  time: 0.6706  data: 0.3351  max mem: 21002
Epoch: [272]  [ 60/312]  eta: 0:02:50  lr: 0.000098  min_lr: 0.000098  loss: 3.7768 (3.6636)  weight_decay: 0.0500 (0.0500)  time: 0.5086  data: 0.1728  max mem: 21002
Epoch: [272]  [ 70/312]  eta: 0:02:44  lr: 0.000097  min_lr: 0.000097  loss: 3.5571 (3.6152)  weight_decay: 0.0500 (0.0500)  time: 0.5029  data: 0.1558  max mem: 21002
Epoch: [272]  [ 80/312]  eta: 0:02:33  lr: 0.000097  min_lr: 0.000097  loss: 3.2743 (3.5863)  weight_decay: 0.0500 (0.0500)  time: 0.6128  data: 0.2380  max mem: 21002
Epoch: [272]  [ 90/312]  eta: 0:02:20  lr: 0.000097  min_lr: 0.000097  loss: 3.2743 (3.5566)  weight_decay: 0.0500 (0.0500)  time: 0.4700  data: 0.1392  max mem: 21002
Epoch: [272]  [100/312]  eta: 0:02:16  lr: 0.000097  min_lr: 0.000097  loss: 3.2780 (3.5247)  weight_decay: 0.0500 (0.0500)  time: 0.5722  data: 0.2016  max mem: 21002
Epoch: [272]  [110/312]  eta: 0:02:03  lr: 0.000096  min_lr: 0.000096  loss: 3.7119 (3.5557)  weight_decay: 0.0500 (0.0500)  time: 0.5156  data: 0.1605  max mem: 21002
Epoch: [272]  [120/312]  eta: 0:01:58  lr: 0.000096  min_lr: 0.000096  loss: 3.8299 (3.5784)  weight_decay: 0.0500 (0.0500)  time: 0.4785  data: 0.1166  max mem: 21002
Epoch: [272]  [130/312]  eta: 0:01:53  lr: 0.000096  min_lr: 0.000096  loss: 3.8185 (3.5831)  weight_decay: 0.0500 (0.0500)  time: 0.6755  data: 0.2051  max mem: 21002
Epoch: [272]  [140/312]  eta: 0:01:43  lr: 0.000096  min_lr: 0.000096  loss: 3.6550 (3.5753)  weight_decay: 0.0500 (0.0500)  time: 0.5107  data: 0.1144  max mem: 21002
Epoch: [272]  [150/312]  eta: 0:01:36  lr: 0.000096  min_lr: 0.000096  loss: 3.6550 (3.5692)  weight_decay: 0.0500 (0.0500)  time: 0.4265  data: 0.1400  max mem: 21002
Epoch: [272]  [160/312]  eta: 0:01:30  lr: 0.000095  min_lr: 0.000095  loss: 3.5166 (3.5655)  weight_decay: 0.0500 (0.0500)  time: 0.5616  data: 0.2712  max mem: 21002
Epoch: [272]  [170/312]  eta: 0:01:22  lr: 0.000095  min_lr: 0.000095  loss: 3.5166 (3.5594)  weight_decay: 0.0500 (0.0500)  time: 0.4457  data: 0.1571  max mem: 21002
Epoch: [272]  [180/312]  eta: 0:01:17  lr: 0.000095  min_lr: 0.000095  loss: 3.8815 (3.5770)  weight_decay: 0.0500 (0.0500)  time: 0.5004  data: 0.1742  max mem: 21002
Epoch: [272]  [190/312]  eta: 0:01:09  lr: 0.000095  min_lr: 0.000095  loss: 3.8416 (3.5748)  weight_decay: 0.0500 (0.0500)  time: 0.5015  data: 0.1742  max mem: 21002
Epoch: [272]  [200/312]  eta: 0:01:04  lr: 0.000094  min_lr: 0.000094  loss: 3.5034 (3.5620)  weight_decay: 0.0500 (0.0500)  time: 0.5261  data: 0.1453  max mem: 21002
Epoch: [272]  [210/312]  eta: 0:00:59  lr: 0.000094  min_lr: 0.000094  loss: 3.2230 (3.5553)  weight_decay: 0.0500 (0.0500)  time: 0.6866  data: 0.2206  max mem: 21002
Epoch: [272]  [220/312]  eta: 0:00:52  lr: 0.000094  min_lr: 0.000094  loss: 3.6349 (3.5634)  weight_decay: 0.0500 (0.0500)  time: 0.4782  data: 0.1028  max mem: 21002
Epoch: [272]  [230/312]  eta: 0:00:47  lr: 0.000094  min_lr: 0.000094  loss: 3.6349 (3.5632)  weight_decay: 0.0500 (0.0500)  time: 0.5318  data: 0.1198  max mem: 21002
Epoch: [272]  [240/312]  eta: 0:00:41  lr: 0.000094  min_lr: 0.000094  loss: 3.4475 (3.5650)  weight_decay: 0.0500 (0.0500)  time: 0.6242  data: 0.1102  max mem: 21002
Epoch: [272]  [250/312]  eta: 0:00:35  lr: 0.000093  min_lr: 0.000093  loss: 3.7223 (3.5741)  weight_decay: 0.0500 (0.0500)  time: 0.5731  data: 0.0863  max mem: 21002
Epoch: [272]  [260/312]  eta: 0:00:29  lr: 0.000093  min_lr: 0.000093  loss: 3.7368 (3.5734)  weight_decay: 0.0500 (0.0500)  time: 0.5361  data: 0.0859  max mem: 21002
Epoch: [272]  [270/312]  eta: 0:00:23  lr: 0.000093  min_lr: 0.000093  loss: 3.7684 (3.5709)  weight_decay: 0.0500 (0.0500)  time: 0.3720  data: 0.0174  max mem: 21002
Epoch: [272]  [280/312]  eta: 0:00:18  lr: 0.000093  min_lr: 0.000093  loss: 3.7293 (3.5707)  weight_decay: 0.0500 (0.0500)  time: 0.5185  data: 0.0173  max mem: 21002
Epoch: [272]  [290/312]  eta: 0:00:12  lr: 0.000093  min_lr: 0.000093  loss: 3.5957 (3.5659)  weight_decay: 0.0500 (0.0500)  time: 0.6092  data: 0.0354  max mem: 21002
Epoch: [272]  [300/312]  eta: 0:00:06  lr: 0.000092  min_lr: 0.000092  loss: 3.5957 (3.5656)  weight_decay: 0.0500 (0.0500)  time: 0.3726  data: 0.0185  max mem: 21002
Epoch: [272]  [310/312]  eta: 0:00:01  lr: 0.000092  min_lr: 0.000092  loss: 3.8155 (3.5674)  weight_decay: 0.0500 (0.0500)  time: 0.2774  data: 0.0001  max mem: 21002
Epoch: [272]  [311/312]  eta: 0:00:00  lr: 0.000092  min_lr: 0.000092  loss: 3.8155 (3.5682)  weight_decay: 0.0500 (0.0500)  time: 0.2774  data: 0.0001  max mem: 21002
Epoch: [272] Total time: 0:02:51 (0.5489 s / it)
Averaged stats: lr: 0.000092  min_lr: 0.000092  loss: 3.8155 (3.5982)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.0507 (1.0507)  acc1: 80.9896 (80.9896)  acc5: 95.9635 (95.9635)  time: 8.7468  data: 8.6281  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3920 (1.2974)  acc1: 72.6562 (73.9040)  acc5: 91.4062 (92.0800)  time: 1.0579  data: 0.9587  max mem: 21002
Test: Total time: 0:00:09 (1.0894 s / it)
* Acc@1 73.558 Acc@5 91.952 loss 1.303
Accuracy of the model on the 50000 test images: 73.6%
Max accuracy: 73.62%
Epoch: [273]  [  0/312]  eta: 0:58:04  lr: 0.000092  min_lr: 0.000092  loss: 2.5531 (2.5531)  weight_decay: 0.0500 (0.0500)  time: 11.1680  data: 9.1910  max mem: 21002
Epoch: [273]  [ 10/312]  eta: 0:07:52  lr: 0.000092  min_lr: 0.000092  loss: 3.1952 (3.2337)  weight_decay: 0.0500 (0.0500)  time: 1.5653  data: 0.9930  max mem: 21002
Epoch: [273]  [ 20/312]  eta: 0:05:07  lr: 0.000092  min_lr: 0.000092  loss: 3.4745 (3.3875)  weight_decay: 0.0500 (0.0500)  time: 0.5466  data: 0.0870  max mem: 21002
Epoch: [273]  [ 30/312]  eta: 0:03:47  lr: 0.000091  min_lr: 0.000091  loss: 3.6604 (3.4974)  weight_decay: 0.0500 (0.0500)  time: 0.3919  data: 0.0059  max mem: 21002
Epoch: [273]  [ 40/312]  eta: 0:03:24  lr: 0.000091  min_lr: 0.000091  loss: 3.8706 (3.5433)  weight_decay: 0.0500 (0.0500)  time: 0.4336  data: 0.0623  max mem: 21002
Epoch: [273]  [ 50/312]  eta: 0:03:12  lr: 0.000091  min_lr: 0.000091  loss: 3.6635 (3.5402)  weight_decay: 0.0500 (0.0500)  time: 0.6172  data: 0.1856  max mem: 21002
Epoch: [273]  [ 60/312]  eta: 0:02:46  lr: 0.000091  min_lr: 0.000091  loss: 3.5761 (3.5170)  weight_decay: 0.0500 (0.0500)  time: 0.4744  data: 0.1293  max mem: 21002
Epoch: [273]  [ 70/312]  eta: 0:02:41  lr: 0.000091  min_lr: 0.000091  loss: 3.7336 (3.5321)  weight_decay: 0.0500 (0.0500)  time: 0.4949  data: 0.1016  max mem: 21002
Epoch: [273]  [ 80/312]  eta: 0:02:34  lr: 0.000090  min_lr: 0.000090  loss: 3.7877 (3.5493)  weight_decay: 0.0500 (0.0500)  time: 0.6763  data: 0.1732  max mem: 21002
Epoch: [273]  [ 90/312]  eta: 0:02:18  lr: 0.000090  min_lr: 0.000090  loss: 3.8722 (3.5616)  weight_decay: 0.0500 (0.0500)  time: 0.4814  data: 0.0779  max mem: 21002
Epoch: [273]  [100/312]  eta: 0:02:13  lr: 0.000090  min_lr: 0.000090  loss: 3.8722 (3.5797)  weight_decay: 0.0500 (0.0500)  time: 0.4944  data: 0.0652  max mem: 21002
Epoch: [273]  [110/312]  eta: 0:02:01  lr: 0.000090  min_lr: 0.000090  loss: 3.7060 (3.5641)  weight_decay: 0.0500 (0.0500)  time: 0.4850  data: 0.0649  max mem: 21002
Epoch: [273]  [120/312]  eta: 0:01:56  lr: 0.000089  min_lr: 0.000089  loss: 3.5591 (3.5521)  weight_decay: 0.0500 (0.0500)  time: 0.4943  data: 0.0294  max mem: 21002
Epoch: [273]  [130/312]  eta: 0:01:51  lr: 0.000089  min_lr: 0.000089  loss: 3.5302 (3.5613)  weight_decay: 0.0500 (0.0500)  time: 0.6954  data: 0.0315  max mem: 21002
Epoch: [273]  [140/312]  eta: 0:01:41  lr: 0.000089  min_lr: 0.000089  loss: 3.5416 (3.5561)  weight_decay: 0.0500 (0.0500)  time: 0.4963  data: 0.0135  max mem: 21002
Epoch: [273]  [150/312]  eta: 0:01:37  lr: 0.000089  min_lr: 0.000089  loss: 3.5846 (3.5602)  weight_decay: 0.0500 (0.0500)  time: 0.5089  data: 0.0297  max mem: 21002
Epoch: [273]  [160/312]  eta: 0:01:31  lr: 0.000089  min_lr: 0.000089  loss: 3.6440 (3.5564)  weight_decay: 0.0500 (0.0500)  time: 0.6814  data: 0.0435  max mem: 21002
Epoch: [273]  [170/312]  eta: 0:01:24  lr: 0.000088  min_lr: 0.000088  loss: 3.4204 (3.5506)  weight_decay: 0.0500 (0.0500)  time: 0.5281  data: 0.0504  max mem: 21002
Epoch: [273]  [180/312]  eta: 0:01:18  lr: 0.000088  min_lr: 0.000088  loss: 3.6235 (3.5614)  weight_decay: 0.0500 (0.0500)  time: 0.5436  data: 0.0654  max mem: 21002
Epoch: [273]  [190/312]  eta: 0:01:10  lr: 0.000088  min_lr: 0.000088  loss: 3.6235 (3.5507)  weight_decay: 0.0500 (0.0500)  time: 0.4850  data: 0.0358  max mem: 21002
Epoch: [273]  [200/312]  eta: 0:01:05  lr: 0.000088  min_lr: 0.000088  loss: 3.4651 (3.5594)  weight_decay: 0.0500 (0.0500)  time: 0.4796  data: 0.0093  max mem: 21002
Epoch: [273]  [210/312]  eta: 0:00:59  lr: 0.000088  min_lr: 0.000088  loss: 3.8447 (3.5664)  weight_decay: 0.0500 (0.0500)  time: 0.6502  data: 0.0154  max mem: 21002
Epoch: [273]  [220/312]  eta: 0:00:53  lr: 0.000087  min_lr: 0.000087  loss: 3.5337 (3.5469)  weight_decay: 0.0500 (0.0500)  time: 0.4959  data: 0.0417  max mem: 21002
Epoch: [273]  [230/312]  eta: 0:00:47  lr: 0.000087  min_lr: 0.000087  loss: 3.3498 (3.5488)  weight_decay: 0.0500 (0.0500)  time: 0.5364  data: 0.0810  max mem: 21002
Epoch: [273]  [240/312]  eta: 0:00:41  lr: 0.000087  min_lr: 0.000087  loss: 3.7171 (3.5512)  weight_decay: 0.0500 (0.0500)  time: 0.6405  data: 0.0470  max mem: 21002
Epoch: [273]  [250/312]  eta: 0:00:35  lr: 0.000087  min_lr: 0.000087  loss: 3.7158 (3.5578)  weight_decay: 0.0500 (0.0500)  time: 0.4721  data: 0.0241  max mem: 21002
Epoch: [273]  [260/312]  eta: 0:00:29  lr: 0.000087  min_lr: 0.000087  loss: 3.6548 (3.5510)  weight_decay: 0.0500 (0.0500)  time: 0.4782  data: 0.0403  max mem: 21002
Epoch: [273]  [270/312]  eta: 0:00:23  lr: 0.000086  min_lr: 0.000086  loss: 3.6907 (3.5524)  weight_decay: 0.0500 (0.0500)  time: 0.4503  data: 0.0365  max mem: 21002
Epoch: [273]  [280/312]  eta: 0:00:18  lr: 0.000086  min_lr: 0.000086  loss: 3.4963 (3.5497)  weight_decay: 0.0500 (0.0500)  time: 0.5228  data: 0.0681  max mem: 21002
Epoch: [273]  [290/312]  eta: 0:00:12  lr: 0.000086  min_lr: 0.000086  loss: 3.4504 (3.5514)  weight_decay: 0.0500 (0.0500)  time: 0.5944  data: 0.0634  max mem: 21002
Epoch: [273]  [300/312]  eta: 0:00:06  lr: 0.000086  min_lr: 0.000086  loss: 3.7381 (3.5455)  weight_decay: 0.0500 (0.0500)  time: 0.3751  data: 0.0154  max mem: 21002
Epoch: [273]  [310/312]  eta: 0:00:01  lr: 0.000086  min_lr: 0.000086  loss: 3.5818 (3.5423)  weight_decay: 0.0500 (0.0500)  time: 0.2775  data: 0.0001  max mem: 21002
Epoch: [273]  [311/312]  eta: 0:00:00  lr: 0.000085  min_lr: 0.000085  loss: 3.3585 (3.5393)  weight_decay: 0.0500 (0.0500)  time: 0.2773  data: 0.0001  max mem: 21002
Epoch: [273] Total time: 0:02:52 (0.5514 s / it)
Averaged stats: lr: 0.000085  min_lr: 0.000085  loss: 3.3585 (3.5915)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.0087 (1.0087)  acc1: 81.6406 (81.6406)  acc5: 96.2240 (96.2240)  time: 8.8532  data: 8.7347  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3499 (1.2584)  acc1: 72.1354 (73.7120)  acc5: 91.5365 (92.0000)  time: 1.0687  data: 0.9706  max mem: 21002
Test: Total time: 0:00:09 (1.0827 s / it)
* Acc@1 73.692 Acc@5 92.050 loss 1.261
Accuracy of the model on the 50000 test images: 73.7%
Max accuracy: 73.69%
Epoch: [274]  [  0/312]  eta: 1:04:40  lr: 0.000085  min_lr: 0.000085  loss: 4.2487 (4.2487)  weight_decay: 0.0500 (0.0500)  time: 12.4381  data: 8.7398  max mem: 21002
Epoch: [274]  [ 10/312]  eta: 0:08:32  lr: 0.000085  min_lr: 0.000085  loss: 3.9121 (3.7245)  weight_decay: 0.0500 (0.0500)  time: 1.6969  data: 1.0796  max mem: 21002
Epoch: [274]  [ 20/312]  eta: 0:05:09  lr: 0.000085  min_lr: 0.000085  loss: 3.7268 (3.6307)  weight_decay: 0.0500 (0.0500)  time: 0.4897  data: 0.1573  max mem: 21002
Epoch: [274]  [ 30/312]  eta: 0:03:48  lr: 0.000085  min_lr: 0.000085  loss: 3.4955 (3.5972)  weight_decay: 0.0500 (0.0500)  time: 0.3234  data: 0.0010  max mem: 21002
Epoch: [274]  [ 40/312]  eta: 0:03:21  lr: 0.000085  min_lr: 0.000085  loss: 3.6564 (3.6445)  weight_decay: 0.0500 (0.0500)  time: 0.4061  data: 0.0767  max mem: 21002
Epoch: [274]  [ 50/312]  eta: 0:03:10  lr: 0.000084  min_lr: 0.000084  loss: 3.8311 (3.6344)  weight_decay: 0.0500 (0.0500)  time: 0.5997  data: 0.1887  max mem: 21002
Epoch: [274]  [ 60/312]  eta: 0:02:46  lr: 0.000084  min_lr: 0.000084  loss: 3.8540 (3.6462)  weight_decay: 0.0500 (0.0500)  time: 0.4947  data: 0.1230  max mem: 21002
Epoch: [274]  [ 70/312]  eta: 0:02:37  lr: 0.000084  min_lr: 0.000084  loss: 3.8540 (3.6415)  weight_decay: 0.0500 (0.0500)  time: 0.4587  data: 0.0963  max mem: 21002
Epoch: [274]  [ 80/312]  eta: 0:02:29  lr: 0.000084  min_lr: 0.000084  loss: 3.8069 (3.6543)  weight_decay: 0.0500 (0.0500)  time: 0.5895  data: 0.1477  max mem: 21002
Epoch: [274]  [ 90/312]  eta: 0:02:16  lr: 0.000084  min_lr: 0.000084  loss: 3.8495 (3.6777)  weight_decay: 0.0500 (0.0500)  time: 0.4885  data: 0.1205  max mem: 21002
Epoch: [274]  [100/312]  eta: 0:02:10  lr: 0.000083  min_lr: 0.000083  loss: 3.9220 (3.6844)  weight_decay: 0.0500 (0.0500)  time: 0.5130  data: 0.1861  max mem: 21002
Epoch: [274]  [110/312]  eta: 0:01:59  lr: 0.000083  min_lr: 0.000083  loss: 3.8825 (3.6890)  weight_decay: 0.0500 (0.0500)  time: 0.4812  data: 0.1494  max mem: 21002
Epoch: [274]  [120/312]  eta: 0:01:56  lr: 0.000083  min_lr: 0.000083  loss: 3.6642 (3.6734)  weight_decay: 0.0500 (0.0500)  time: 0.5411  data: 0.1521  max mem: 21002
Epoch: [274]  [130/312]  eta: 0:01:50  lr: 0.000083  min_lr: 0.000083  loss: 3.5529 (3.6514)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.2163  max mem: 21002
Epoch: [274]  [140/312]  eta: 0:01:41  lr: 0.000083  min_lr: 0.000083  loss: 3.4497 (3.6297)  weight_decay: 0.0500 (0.0500)  time: 0.4873  data: 0.1185  max mem: 21002
Epoch: [274]  [150/312]  eta: 0:01:35  lr: 0.000082  min_lr: 0.000082  loss: 3.4511 (3.6277)  weight_decay: 0.0500 (0.0500)  time: 0.4789  data: 0.1165  max mem: 21002
Epoch: [274]  [160/312]  eta: 0:01:28  lr: 0.000082  min_lr: 0.000082  loss: 3.7196 (3.6303)  weight_decay: 0.0500 (0.0500)  time: 0.5609  data: 0.1599  max mem: 21002
Epoch: [274]  [170/312]  eta: 0:01:21  lr: 0.000082  min_lr: 0.000082  loss: 3.6400 (3.6185)  weight_decay: 0.0500 (0.0500)  time: 0.4882  data: 0.1313  max mem: 21002
Epoch: [274]  [180/312]  eta: 0:01:17  lr: 0.000082  min_lr: 0.000082  loss: 3.6820 (3.6219)  weight_decay: 0.0500 (0.0500)  time: 0.5785  data: 0.1756  max mem: 21002
Epoch: [274]  [190/312]  eta: 0:01:09  lr: 0.000082  min_lr: 0.000082  loss: 3.5680 (3.6132)  weight_decay: 0.0500 (0.0500)  time: 0.4975  data: 0.1295  max mem: 21002
Epoch: [274]  [200/312]  eta: 0:01:04  lr: 0.000081  min_lr: 0.000081  loss: 3.7111 (3.6213)  weight_decay: 0.0500 (0.0500)  time: 0.4873  data: 0.1419  max mem: 21002
Epoch: [274]  [210/312]  eta: 0:00:59  lr: 0.000081  min_lr: 0.000081  loss: 3.7863 (3.6243)  weight_decay: 0.0500 (0.0500)  time: 0.7083  data: 0.2693  max mem: 21002
Epoch: [274]  [220/312]  eta: 0:00:52  lr: 0.000081  min_lr: 0.000081  loss: 3.5270 (3.6137)  weight_decay: 0.0500 (0.0500)  time: 0.5375  data: 0.1564  max mem: 21002
Epoch: [274]  [230/312]  eta: 0:00:47  lr: 0.000081  min_lr: 0.000081  loss: 3.3845 (3.6114)  weight_decay: 0.0500 (0.0500)  time: 0.4867  data: 0.1133  max mem: 21002
Epoch: [274]  [240/312]  eta: 0:00:41  lr: 0.000081  min_lr: 0.000081  loss: 3.7130 (3.6139)  weight_decay: 0.0500 (0.0500)  time: 0.6309  data: 0.2149  max mem: 21002
Epoch: [274]  [250/312]  eta: 0:00:35  lr: 0.000080  min_lr: 0.000080  loss: 3.7347 (3.6114)  weight_decay: 0.0500 (0.0500)  time: 0.4930  data: 0.1214  max mem: 21002
Epoch: [274]  [260/312]  eta: 0:00:29  lr: 0.000080  min_lr: 0.000080  loss: 3.3360 (3.5987)  weight_decay: 0.0500 (0.0500)  time: 0.5155  data: 0.1161  max mem: 21002
Epoch: [274]  [270/312]  eta: 0:00:23  lr: 0.000080  min_lr: 0.000080  loss: 3.5155 (3.6015)  weight_decay: 0.0500 (0.0500)  time: 0.4781  data: 0.1169  max mem: 21002
Epoch: [274]  [280/312]  eta: 0:00:18  lr: 0.000080  min_lr: 0.000080  loss: 3.8285 (3.6027)  weight_decay: 0.0500 (0.0500)  time: 0.4846  data: 0.1157  max mem: 21002
Epoch: [274]  [290/312]  eta: 0:00:12  lr: 0.000080  min_lr: 0.000080  loss: 3.6660 (3.6031)  weight_decay: 0.0500 (0.0500)  time: 0.6407  data: 0.2263  max mem: 21002
Epoch: [274]  [300/312]  eta: 0:00:06  lr: 0.000079  min_lr: 0.000079  loss: 3.5425 (3.5960)  weight_decay: 0.0500 (0.0500)  time: 0.4418  data: 0.1118  max mem: 21002
Epoch: [274]  [310/312]  eta: 0:00:01  lr: 0.000079  min_lr: 0.000079  loss: 3.7263 (3.6020)  weight_decay: 0.0500 (0.0500)  time: 0.3046  data: 0.0102  max mem: 21002
Epoch: [274]  [311/312]  eta: 0:00:00  lr: 0.000079  min_lr: 0.000079  loss: 3.7254 (3.5999)  weight_decay: 0.0500 (0.0500)  time: 0.3045  data: 0.0102  max mem: 21002
Epoch: [274] Total time: 0:02:52 (0.5526 s / it)
Averaged stats: lr: 0.000079  min_lr: 0.000079  loss: 3.7254 (3.5718)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.0535 (1.0535)  acc1: 80.3385 (80.3385)  acc5: 95.0521 (95.0521)  time: 8.4903  data: 8.3722  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3928 (1.3019)  acc1: 72.5260 (73.4240)  acc5: 91.0156 (91.5520)  time: 1.0298  data: 0.9303  max mem: 21002
Test: Total time: 0:00:09 (1.0428 s / it)
* Acc@1 73.536 Acc@5 91.844 loss 1.308
Accuracy of the model on the 50000 test images: 73.5%
Max accuracy: 73.69%
Epoch: [275]  [  0/312]  eta: 1:09:15  lr: 0.000079  min_lr: 0.000079  loss: 2.9091 (2.9091)  weight_decay: 0.0500 (0.0500)  time: 13.3183  data: 9.2037  max mem: 21002
Epoch: [275]  [ 10/312]  eta: 0:08:48  lr: 0.000079  min_lr: 0.000079  loss: 3.8147 (3.8857)  weight_decay: 0.0500 (0.0500)  time: 1.7496  data: 1.0506  max mem: 21002
Epoch: [275]  [ 20/312]  eta: 0:05:23  lr: 0.000079  min_lr: 0.000079  loss: 3.6633 (3.5808)  weight_decay: 0.0500 (0.0500)  time: 0.4975  data: 0.1296  max mem: 21002
Epoch: [275]  [ 30/312]  eta: 0:03:58  lr: 0.000079  min_lr: 0.000079  loss: 3.4891 (3.5794)  weight_decay: 0.0500 (0.0500)  time: 0.3474  data: 0.0127  max mem: 21002
Epoch: [275]  [ 40/312]  eta: 0:03:13  lr: 0.000078  min_lr: 0.000078  loss: 3.5190 (3.5647)  weight_decay: 0.0500 (0.0500)  time: 0.2922  data: 0.0020  max mem: 21002
Epoch: [275]  [ 50/312]  eta: 0:03:03  lr: 0.000078  min_lr: 0.000078  loss: 3.5190 (3.5656)  weight_decay: 0.0500 (0.0500)  time: 0.4729  data: 0.0105  max mem: 21002
Epoch: [275]  [ 60/312]  eta: 0:02:43  lr: 0.000078  min_lr: 0.000078  loss: 3.4303 (3.5113)  weight_decay: 0.0500 (0.0500)  time: 0.5263  data: 0.0658  max mem: 21002
Epoch: [275]  [ 70/312]  eta: 0:02:36  lr: 0.000078  min_lr: 0.000078  loss: 3.6753 (3.5617)  weight_decay: 0.0500 (0.0500)  time: 0.5060  data: 0.0929  max mem: 21002
Epoch: [275]  [ 80/312]  eta: 0:02:30  lr: 0.000078  min_lr: 0.000078  loss: 3.8081 (3.5524)  weight_decay: 0.0500 (0.0500)  time: 0.6492  data: 0.0873  max mem: 21002
Epoch: [275]  [ 90/312]  eta: 0:02:18  lr: 0.000077  min_lr: 0.000077  loss: 3.7789 (3.5611)  weight_decay: 0.0500 (0.0500)  time: 0.5462  data: 0.1127  max mem: 21002
Epoch: [275]  [100/312]  eta: 0:02:13  lr: 0.000077  min_lr: 0.000077  loss: 3.6976 (3.5749)  weight_decay: 0.0500 (0.0500)  time: 0.5476  data: 0.1152  max mem: 21002
Epoch: [275]  [110/312]  eta: 0:02:01  lr: 0.000077  min_lr: 0.000077  loss: 3.5984 (3.5651)  weight_decay: 0.0500 (0.0500)  time: 0.5042  data: 0.0720  max mem: 21002
Epoch: [275]  [120/312]  eta: 0:01:55  lr: 0.000077  min_lr: 0.000077  loss: 3.5499 (3.5531)  weight_decay: 0.0500 (0.0500)  time: 0.4635  data: 0.0480  max mem: 21002
Epoch: [275]  [130/312]  eta: 0:01:50  lr: 0.000077  min_lr: 0.000077  loss: 3.6477 (3.5576)  weight_decay: 0.0500 (0.0500)  time: 0.6392  data: 0.0826  max mem: 21002
Epoch: [275]  [140/312]  eta: 0:01:41  lr: 0.000076  min_lr: 0.000076  loss: 3.4478 (3.5455)  weight_decay: 0.0500 (0.0500)  time: 0.5207  data: 0.0996  max mem: 21002
Epoch: [275]  [150/312]  eta: 0:01:36  lr: 0.000076  min_lr: 0.000076  loss: 3.6779 (3.5643)  weight_decay: 0.0500 (0.0500)  time: 0.5200  data: 0.1018  max mem: 21002
Epoch: [275]  [160/312]  eta: 0:01:30  lr: 0.000076  min_lr: 0.000076  loss: 3.7341 (3.5551)  weight_decay: 0.0500 (0.0500)  time: 0.6425  data: 0.1061  max mem: 21002
Epoch: [275]  [170/312]  eta: 0:01:22  lr: 0.000076  min_lr: 0.000076  loss: 3.6472 (3.5534)  weight_decay: 0.0500 (0.0500)  time: 0.4906  data: 0.0910  max mem: 21002
Epoch: [275]  [180/312]  eta: 0:01:17  lr: 0.000076  min_lr: 0.000076  loss: 3.8077 (3.5693)  weight_decay: 0.0500 (0.0500)  time: 0.5177  data: 0.0942  max mem: 21002
Epoch: [275]  [190/312]  eta: 0:01:10  lr: 0.000075  min_lr: 0.000075  loss: 3.7524 (3.5777)  weight_decay: 0.0500 (0.0500)  time: 0.4980  data: 0.0755  max mem: 21002
Epoch: [275]  [200/312]  eta: 0:01:04  lr: 0.000075  min_lr: 0.000075  loss: 3.6552 (3.5824)  weight_decay: 0.0500 (0.0500)  time: 0.4745  data: 0.0493  max mem: 21002
Epoch: [275]  [210/312]  eta: 0:00:59  lr: 0.000075  min_lr: 0.000075  loss: 3.4174 (3.5712)  weight_decay: 0.0500 (0.0500)  time: 0.6432  data: 0.0894  max mem: 21002
Epoch: [275]  [220/312]  eta: 0:00:52  lr: 0.000075  min_lr: 0.000075  loss: 3.4174 (3.5674)  weight_decay: 0.0500 (0.0500)  time: 0.5486  data: 0.1157  max mem: 21002
Epoch: [275]  [230/312]  eta: 0:00:47  lr: 0.000075  min_lr: 0.000075  loss: 3.6752 (3.5725)  weight_decay: 0.0500 (0.0500)  time: 0.5090  data: 0.0930  max mem: 21002
Epoch: [275]  [240/312]  eta: 0:00:41  lr: 0.000074  min_lr: 0.000074  loss: 3.6752 (3.5680)  weight_decay: 0.0500 (0.0500)  time: 0.6060  data: 0.0893  max mem: 21002
Epoch: [275]  [250/312]  eta: 0:00:35  lr: 0.000074  min_lr: 0.000074  loss: 3.5154 (3.5695)  weight_decay: 0.0500 (0.0500)  time: 0.5588  data: 0.1063  max mem: 21002
Epoch: [275]  [260/312]  eta: 0:00:29  lr: 0.000074  min_lr: 0.000074  loss: 3.6210 (3.5645)  weight_decay: 0.0500 (0.0500)  time: 0.4934  data: 0.0844  max mem: 21002
Epoch: [275]  [270/312]  eta: 0:00:23  lr: 0.000074  min_lr: 0.000074  loss: 3.4873 (3.5527)  weight_decay: 0.0500 (0.0500)  time: 0.4678  data: 0.0704  max mem: 21002
Epoch: [275]  [280/312]  eta: 0:00:18  lr: 0.000074  min_lr: 0.000074  loss: 3.5675 (3.5608)  weight_decay: 0.0500 (0.0500)  time: 0.5269  data: 0.0851  max mem: 21002
Epoch: [275]  [290/312]  eta: 0:00:12  lr: 0.000073  min_lr: 0.000073  loss: 3.8606 (3.5655)  weight_decay: 0.0500 (0.0500)  time: 0.6085  data: 0.1068  max mem: 21002
Epoch: [275]  [300/312]  eta: 0:00:06  lr: 0.000073  min_lr: 0.000073  loss: 3.6521 (3.5648)  weight_decay: 0.0500 (0.0500)  time: 0.4563  data: 0.0724  max mem: 21002
Epoch: [275]  [310/312]  eta: 0:00:01  lr: 0.000073  min_lr: 0.000073  loss: 3.5108 (3.5601)  weight_decay: 0.0500 (0.0500)  time: 0.2934  data: 0.0086  max mem: 21002
Epoch: [275]  [311/312]  eta: 0:00:00  lr: 0.000073  min_lr: 0.000073  loss: 3.5108 (3.5574)  weight_decay: 0.0500 (0.0500)  time: 0.2933  data: 0.0086  max mem: 21002
Epoch: [275] Total time: 0:02:52 (0.5539 s / it)
Averaged stats: lr: 0.000073  min_lr: 0.000073  loss: 3.5108 (3.5804)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.0021 (1.0021)  acc1: 81.2500 (81.2500)  acc5: 95.3125 (95.3125)  time: 8.4605  data: 8.3423  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3356 (1.2570)  acc1: 72.3958 (73.5040)  acc5: 91.5365 (91.8080)  time: 1.0260  data: 0.9270  max mem: 21002
Test: Total time: 0:00:09 (1.0384 s / it)
* Acc@1 73.588 Acc@5 91.954 loss 1.260
Accuracy of the model on the 50000 test images: 73.6%
Max accuracy: 73.69%
Epoch: [276]  [  0/312]  eta: 1:02:19  lr: 0.000073  min_lr: 0.000073  loss: 3.5127 (3.5127)  weight_decay: 0.0500 (0.0500)  time: 11.9851  data: 8.1573  max mem: 21002
Epoch: [276]  [ 10/312]  eta: 0:08:03  lr: 0.000073  min_lr: 0.000073  loss: 3.6729 (3.5423)  weight_decay: 0.0500 (0.0500)  time: 1.6000  data: 0.9964  max mem: 21002
Epoch: [276]  [ 20/312]  eta: 0:05:17  lr: 0.000073  min_lr: 0.000073  loss: 3.6076 (3.5259)  weight_decay: 0.0500 (0.0500)  time: 0.5411  data: 0.1405  max mem: 21002
Epoch: [276]  [ 30/312]  eta: 0:03:54  lr: 0.000072  min_lr: 0.000072  loss: 3.6076 (3.5211)  weight_decay: 0.0500 (0.0500)  time: 0.4062  data: 0.0008  max mem: 21002
Epoch: [276]  [ 40/312]  eta: 0:03:25  lr: 0.000072  min_lr: 0.000072  loss: 3.7013 (3.5154)  weight_decay: 0.0500 (0.0500)  time: 0.4058  data: 0.0009  max mem: 21002
Epoch: [276]  [ 50/312]  eta: 0:03:18  lr: 0.000072  min_lr: 0.000072  loss: 3.5285 (3.4829)  weight_decay: 0.0500 (0.0500)  time: 0.6414  data: 0.0693  max mem: 21002
Epoch: [276]  [ 60/312]  eta: 0:02:51  lr: 0.000072  min_lr: 0.000072  loss: 3.4464 (3.4896)  weight_decay: 0.0500 (0.0500)  time: 0.5284  data: 0.0693  max mem: 21002
Epoch: [276]  [ 70/312]  eta: 0:02:43  lr: 0.000072  min_lr: 0.000072  loss: 3.6327 (3.5133)  weight_decay: 0.0500 (0.0500)  time: 0.4769  data: 0.0315  max mem: 21002
Epoch: [276]  [ 80/312]  eta: 0:02:35  lr: 0.000072  min_lr: 0.000072  loss: 3.6758 (3.5179)  weight_decay: 0.0500 (0.0500)  time: 0.6419  data: 0.0314  max mem: 21002
Epoch: [276]  [ 90/312]  eta: 0:02:20  lr: 0.000071  min_lr: 0.000071  loss: 3.5814 (3.5091)  weight_decay: 0.0500 (0.0500)  time: 0.4841  data: 0.0301  max mem: 21002
Epoch: [276]  [100/312]  eta: 0:02:14  lr: 0.000071  min_lr: 0.000071  loss: 3.5814 (3.5175)  weight_decay: 0.0500 (0.0500)  time: 0.4944  data: 0.0632  max mem: 21002
Epoch: [276]  [110/312]  eta: 0:02:02  lr: 0.000071  min_lr: 0.000071  loss: 3.7151 (3.5282)  weight_decay: 0.0500 (0.0500)  time: 0.4655  data: 0.0343  max mem: 21002
Epoch: [276]  [120/312]  eta: 0:01:57  lr: 0.000071  min_lr: 0.000071  loss: 3.8195 (3.5496)  weight_decay: 0.0500 (0.0500)  time: 0.4818  data: 0.0191  max mem: 21002
Epoch: [276]  [130/312]  eta: 0:01:52  lr: 0.000071  min_lr: 0.000071  loss: 3.6086 (3.5479)  weight_decay: 0.0500 (0.0500)  time: 0.6998  data: 0.0485  max mem: 21002
Epoch: [276]  [140/312]  eta: 0:01:42  lr: 0.000070  min_lr: 0.000070  loss: 3.5655 (3.5465)  weight_decay: 0.0500 (0.0500)  time: 0.5061  data: 0.0315  max mem: 21002
Epoch: [276]  [150/312]  eta: 0:01:37  lr: 0.000070  min_lr: 0.000070  loss: 3.6132 (3.5562)  weight_decay: 0.0500 (0.0500)  time: 0.5050  data: 0.0177  max mem: 21002
Epoch: [276]  [160/312]  eta: 0:01:32  lr: 0.000070  min_lr: 0.000070  loss: 3.6443 (3.5560)  weight_decay: 0.0500 (0.0500)  time: 0.6803  data: 0.0170  max mem: 21002
Epoch: [276]  [170/312]  eta: 0:01:24  lr: 0.000070  min_lr: 0.000070  loss: 3.7344 (3.5700)  weight_decay: 0.0500 (0.0500)  time: 0.5013  data: 0.0370  max mem: 21002
Epoch: [276]  [180/312]  eta: 0:01:18  lr: 0.000070  min_lr: 0.000070  loss: 3.7738 (3.5598)  weight_decay: 0.0500 (0.0500)  time: 0.5141  data: 0.0668  max mem: 21002
Epoch: [276]  [190/312]  eta: 0:01:10  lr: 0.000069  min_lr: 0.000069  loss: 3.7485 (3.5699)  weight_decay: 0.0500 (0.0500)  time: 0.4787  data: 0.0306  max mem: 21002
Epoch: [276]  [200/312]  eta: 0:01:05  lr: 0.000069  min_lr: 0.000069  loss: 3.7869 (3.5703)  weight_decay: 0.0500 (0.0500)  time: 0.4807  data: 0.0292  max mem: 21002
Epoch: [276]  [210/312]  eta: 0:01:00  lr: 0.000069  min_lr: 0.000069  loss: 3.7818 (3.5754)  weight_decay: 0.0500 (0.0500)  time: 0.6792  data: 0.0624  max mem: 21002
Epoch: [276]  [220/312]  eta: 0:00:52  lr: 0.000069  min_lr: 0.000069  loss: 3.7789 (3.5867)  weight_decay: 0.0500 (0.0500)  time: 0.4884  data: 0.0339  max mem: 21002
Epoch: [276]  [230/312]  eta: 0:00:47  lr: 0.000069  min_lr: 0.000069  loss: 3.9038 (3.6042)  weight_decay: 0.0500 (0.0500)  time: 0.4940  data: 0.0455  max mem: 21002
Epoch: [276]  [240/312]  eta: 0:00:41  lr: 0.000069  min_lr: 0.000069  loss: 3.8317 (3.5990)  weight_decay: 0.0500 (0.0500)  time: 0.6661  data: 0.0504  max mem: 21002
Epoch: [276]  [250/312]  eta: 0:00:35  lr: 0.000068  min_lr: 0.000068  loss: 3.6445 (3.5971)  weight_decay: 0.0500 (0.0500)  time: 0.5159  data: 0.0312  max mem: 21002
Epoch: [276]  [260/312]  eta: 0:00:30  lr: 0.000068  min_lr: 0.000068  loss: 3.6800 (3.5963)  weight_decay: 0.0500 (0.0500)  time: 0.5381  data: 0.0368  max mem: 21002
Epoch: [276]  [270/312]  eta: 0:00:23  lr: 0.000068  min_lr: 0.000068  loss: 3.6800 (3.6046)  weight_decay: 0.0500 (0.0500)  time: 0.4881  data: 0.0159  max mem: 21002
Epoch: [276]  [280/312]  eta: 0:00:18  lr: 0.000068  min_lr: 0.000068  loss: 3.5942 (3.5975)  weight_decay: 0.0500 (0.0500)  time: 0.5581  data: 0.0408  max mem: 21002
Epoch: [276]  [290/312]  eta: 0:00:12  lr: 0.000068  min_lr: 0.000068  loss: 3.4945 (3.6002)  weight_decay: 0.0500 (0.0500)  time: 0.6087  data: 0.0358  max mem: 21002
Epoch: [276]  [300/312]  eta: 0:00:06  lr: 0.000067  min_lr: 0.000067  loss: 3.7309 (3.6029)  weight_decay: 0.0500 (0.0500)  time: 0.3390  data: 0.0002  max mem: 21002
Epoch: [276]  [310/312]  eta: 0:00:01  lr: 0.000067  min_lr: 0.000067  loss: 3.8712 (3.6080)  weight_decay: 0.0500 (0.0500)  time: 0.2771  data: 0.0001  max mem: 21002
Epoch: [276]  [311/312]  eta: 0:00:00  lr: 0.000067  min_lr: 0.000067  loss: 3.8735 (3.6092)  weight_decay: 0.0500 (0.0500)  time: 0.2770  data: 0.0001  max mem: 21002
Epoch: [276] Total time: 0:02:53 (0.5562 s / it)
Averaged stats: lr: 0.000067  min_lr: 0.000067  loss: 3.8735 (3.5801)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:11  loss: 1.1077 (1.1077)  acc1: 81.3802 (81.3802)  acc5: 96.0938 (96.0938)  time: 7.9802  data: 7.8606  max mem: 21002
Test:  [8/9]  eta: 0:00:00  loss: 1.4208 (1.3437)  acc1: 72.7865 (73.7440)  acc5: 91.4062 (91.9680)  time: 0.9715  data: 0.8735  max mem: 21002
Test: Total time: 0:00:08 (0.9837 s / it)
* Acc@1 73.644 Acc@5 91.958 loss 1.350
Accuracy of the model on the 50000 test images: 73.6%
Max accuracy: 73.69%
Epoch: [277]  [  0/312]  eta: 1:10:11  lr: 0.000067  min_lr: 0.000067  loss: 4.2867 (4.2867)  weight_decay: 0.0500 (0.0500)  time: 13.4978  data: 11.6528  max mem: 21002
Epoch: [277]  [ 10/312]  eta: 0:09:04  lr: 0.000067  min_lr: 0.000067  loss: 3.6758 (3.5229)  weight_decay: 0.0500 (0.0500)  time: 1.8025  data: 1.0598  max mem: 21002
Epoch: [277]  [ 20/312]  eta: 0:05:26  lr: 0.000067  min_lr: 0.000067  loss: 3.6758 (3.6311)  weight_decay: 0.0500 (0.0500)  time: 0.4991  data: 0.0399  max mem: 21002
Epoch: [277]  [ 30/312]  eta: 0:03:59  lr: 0.000067  min_lr: 0.000067  loss: 3.8202 (3.7040)  weight_decay: 0.0500 (0.0500)  time: 0.3275  data: 0.0399  max mem: 21002
Epoch: [277]  [ 40/312]  eta: 0:03:13  lr: 0.000066  min_lr: 0.000066  loss: 3.7169 (3.6393)  weight_decay: 0.0500 (0.0500)  time: 0.2872  data: 0.0006  max mem: 21002
Epoch: [277]  [ 50/312]  eta: 0:02:56  lr: 0.000066  min_lr: 0.000066  loss: 3.7169 (3.6411)  weight_decay: 0.0500 (0.0500)  time: 0.4022  data: 0.0133  max mem: 21002
Epoch: [277]  [ 60/312]  eta: 0:02:40  lr: 0.000066  min_lr: 0.000066  loss: 3.6208 (3.5903)  weight_decay: 0.0500 (0.0500)  time: 0.4865  data: 0.0837  max mem: 21002
Epoch: [277]  [ 70/312]  eta: 0:02:33  lr: 0.000066  min_lr: 0.000066  loss: 3.4157 (3.5458)  weight_decay: 0.0500 (0.0500)  time: 0.5378  data: 0.1067  max mem: 21002
Epoch: [277]  [ 80/312]  eta: 0:02:22  lr: 0.000066  min_lr: 0.000066  loss: 3.4157 (3.5421)  weight_decay: 0.0500 (0.0500)  time: 0.5436  data: 0.1025  max mem: 21002
Epoch: [277]  [ 90/312]  eta: 0:02:16  lr: 0.000066  min_lr: 0.000066  loss: 3.6179 (3.5566)  weight_decay: 0.0500 (0.0500)  time: 0.5406  data: 0.1103  max mem: 21002
Epoch: [277]  [100/312]  eta: 0:02:07  lr: 0.000065  min_lr: 0.000065  loss: 3.8180 (3.5675)  weight_decay: 0.0500 (0.0500)  time: 0.5535  data: 0.0633  max mem: 21002
Epoch: [277]  [110/312]  eta: 0:02:01  lr: 0.000065  min_lr: 0.000065  loss: 3.7205 (3.5596)  weight_decay: 0.0500 (0.0500)  time: 0.5285  data: 0.1079  max mem: 21002
Epoch: [277]  [120/312]  eta: 0:01:53  lr: 0.000065  min_lr: 0.000065  loss: 3.6382 (3.5556)  weight_decay: 0.0500 (0.0500)  time: 0.5398  data: 0.1186  max mem: 21002
Epoch: [277]  [130/312]  eta: 0:01:47  lr: 0.000065  min_lr: 0.000065  loss: 3.6929 (3.5583)  weight_decay: 0.0500 (0.0500)  time: 0.5576  data: 0.0734  max mem: 21002
Epoch: [277]  [140/312]  eta: 0:01:42  lr: 0.000065  min_lr: 0.000065  loss: 3.6227 (3.5461)  weight_decay: 0.0500 (0.0500)  time: 0.6050  data: 0.0950  max mem: 21002
Epoch: [277]  [150/312]  eta: 0:01:35  lr: 0.000064  min_lr: 0.000064  loss: 3.5948 (3.5576)  weight_decay: 0.0500 (0.0500)  time: 0.5674  data: 0.0680  max mem: 21002
Epoch: [277]  [160/312]  eta: 0:01:28  lr: 0.000064  min_lr: 0.000064  loss: 3.8819 (3.5738)  weight_decay: 0.0500 (0.0500)  time: 0.4940  data: 0.0703  max mem: 21002
Epoch: [277]  [170/312]  eta: 0:01:23  lr: 0.000064  min_lr: 0.000064  loss: 3.8513 (3.5870)  weight_decay: 0.0500 (0.0500)  time: 0.5748  data: 0.1080  max mem: 21002
Epoch: [277]  [180/312]  eta: 0:01:16  lr: 0.000064  min_lr: 0.000064  loss: 3.6584 (3.5812)  weight_decay: 0.0500 (0.0500)  time: 0.5903  data: 0.0548  max mem: 21002
Epoch: [277]  [190/312]  eta: 0:01:10  lr: 0.000064  min_lr: 0.000064  loss: 3.5462 (3.5660)  weight_decay: 0.0500 (0.0500)  time: 0.5031  data: 0.0514  max mem: 21002
Epoch: [277]  [200/312]  eta: 0:01:04  lr: 0.000064  min_lr: 0.000064  loss: 3.3250 (3.5613)  weight_decay: 0.0500 (0.0500)  time: 0.5130  data: 0.0510  max mem: 21002
Epoch: [277]  [210/312]  eta: 0:00:59  lr: 0.000063  min_lr: 0.000063  loss: 3.5522 (3.5562)  weight_decay: 0.0500 (0.0500)  time: 0.6048  data: 0.0667  max mem: 21002
Epoch: [277]  [220/312]  eta: 0:00:52  lr: 0.000063  min_lr: 0.000063  loss: 3.5804 (3.5452)  weight_decay: 0.0500 (0.0500)  time: 0.5715  data: 0.0877  max mem: 21002
Epoch: [277]  [230/312]  eta: 0:00:47  lr: 0.000063  min_lr: 0.000063  loss: 3.5804 (3.5478)  weight_decay: 0.0500 (0.0500)  time: 0.5360  data: 0.0714  max mem: 21002
Epoch: [277]  [240/312]  eta: 0:00:41  lr: 0.000063  min_lr: 0.000063  loss: 3.4148 (3.5404)  weight_decay: 0.0500 (0.0500)  time: 0.5146  data: 0.0816  max mem: 21002
Epoch: [277]  [250/312]  eta: 0:00:35  lr: 0.000063  min_lr: 0.000063  loss: 3.3222 (3.5326)  weight_decay: 0.0500 (0.0500)  time: 0.5386  data: 0.1000  max mem: 21002
Epoch: [277]  [260/312]  eta: 0:00:29  lr: 0.000063  min_lr: 0.000063  loss: 3.7514 (3.5425)  weight_decay: 0.0500 (0.0500)  time: 0.6422  data: 0.1059  max mem: 21002
Epoch: [277]  [270/312]  eta: 0:00:23  lr: 0.000062  min_lr: 0.000062  loss: 3.7514 (3.5385)  weight_decay: 0.0500 (0.0500)  time: 0.5043  data: 0.0931  max mem: 21002
Epoch: [277]  [280/312]  eta: 0:00:18  lr: 0.000062  min_lr: 0.000062  loss: 3.6555 (3.5498)  weight_decay: 0.0500 (0.0500)  time: 0.5412  data: 0.1179  max mem: 21002
Epoch: [277]  [290/312]  eta: 0:00:12  lr: 0.000062  min_lr: 0.000062  loss: 3.8642 (3.5474)  weight_decay: 0.0500 (0.0500)  time: 0.5969  data: 0.0918  max mem: 21002
Epoch: [277]  [300/312]  eta: 0:00:06  lr: 0.000062  min_lr: 0.000062  loss: 3.6350 (3.5470)  weight_decay: 0.0500 (0.0500)  time: 0.4128  data: 0.0492  max mem: 21002
Epoch: [277]  [310/312]  eta: 0:00:01  lr: 0.000062  min_lr: 0.000062  loss: 3.6350 (3.5427)  weight_decay: 0.0500 (0.0500)  time: 0.3029  data: 0.0199  max mem: 21002
Epoch: [277]  [311/312]  eta: 0:00:00  lr: 0.000062  min_lr: 0.000062  loss: 3.6287 (3.5414)  weight_decay: 0.0500 (0.0500)  time: 0.3028  data: 0.0199  max mem: 21002
Epoch: [277] Total time: 0:02:53 (0.5571 s / it)
Averaged stats: lr: 0.000062  min_lr: 0.000062  loss: 3.6287 (3.5615)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:08  loss: 1.0385 (1.0385)  acc1: 80.4688 (80.4688)  acc5: 95.8333 (95.8333)  time: 7.6162  data: 7.4969  max mem: 21002
Test:  [8/9]  eta: 0:00:00  loss: 1.3422 (1.2582)  acc1: 73.5677 (73.9680)  acc5: 91.6667 (92.0800)  time: 0.9715  data: 0.8749  max mem: 21002
Test: Total time: 0:00:08 (0.9821 s / it)
* Acc@1 73.650 Acc@5 91.988 loss 1.262
Accuracy of the model on the 50000 test images: 73.7%
Max accuracy: 73.69%
Epoch: [278]  [  0/312]  eta: 1:01:13  lr: 0.000062  min_lr: 0.000062  loss: 3.9545 (3.9545)  weight_decay: 0.0500 (0.0500)  time: 11.7741  data: 9.8301  max mem: 21002
Epoch: [278]  [ 10/312]  eta: 0:08:09  lr: 0.000061  min_lr: 0.000061  loss: 3.7312 (3.6209)  weight_decay: 0.0500 (0.0500)  time: 1.6193  data: 1.0363  max mem: 21002
Epoch: [278]  [ 20/312]  eta: 0:05:20  lr: 0.000061  min_lr: 0.000061  loss: 3.6693 (3.6093)  weight_decay: 0.0500 (0.0500)  time: 0.5644  data: 0.0836  max mem: 21002
Epoch: [278]  [ 30/312]  eta: 0:03:56  lr: 0.000061  min_lr: 0.000061  loss: 3.5884 (3.5880)  weight_decay: 0.0500 (0.0500)  time: 0.4109  data: 0.0055  max mem: 21002
Epoch: [278]  [ 40/312]  eta: 0:03:31  lr: 0.000061  min_lr: 0.000061  loss: 3.5871 (3.5756)  weight_decay: 0.0500 (0.0500)  time: 0.4429  data: 0.0009  max mem: 21002
Epoch: [278]  [ 50/312]  eta: 0:03:20  lr: 0.000061  min_lr: 0.000061  loss: 3.6963 (3.6180)  weight_decay: 0.0500 (0.0500)  time: 0.6470  data: 0.0010  max mem: 21002
Epoch: [278]  [ 60/312]  eta: 0:02:52  lr: 0.000061  min_lr: 0.000061  loss: 3.6301 (3.5804)  weight_decay: 0.0500 (0.0500)  time: 0.4967  data: 0.0010  max mem: 21002
Epoch: [278]  [ 70/312]  eta: 0:02:46  lr: 0.000060  min_lr: 0.000060  loss: 3.7524 (3.6172)  weight_decay: 0.0500 (0.0500)  time: 0.4912  data: 0.0011  max mem: 21002
Epoch: [278]  [ 80/312]  eta: 0:02:36  lr: 0.000060  min_lr: 0.000060  loss: 3.7855 (3.6074)  weight_decay: 0.0500 (0.0500)  time: 0.6397  data: 0.0120  max mem: 21002
Epoch: [278]  [ 90/312]  eta: 0:02:24  lr: 0.000060  min_lr: 0.000060  loss: 3.4855 (3.5902)  weight_decay: 0.0500 (0.0500)  time: 0.5229  data: 0.0119  max mem: 21002
Epoch: [278]  [100/312]  eta: 0:02:16  lr: 0.000060  min_lr: 0.000060  loss: 3.5411 (3.5890)  weight_decay: 0.0500 (0.0500)  time: 0.5248  data: 0.0270  max mem: 21002
Epoch: [278]  [110/312]  eta: 0:02:03  lr: 0.000060  min_lr: 0.000060  loss: 3.5411 (3.5744)  weight_decay: 0.0500 (0.0500)  time: 0.4431  data: 0.0272  max mem: 21002
Epoch: [278]  [120/312]  eta: 0:01:58  lr: 0.000060  min_lr: 0.000060  loss: 3.5210 (3.5654)  weight_decay: 0.0500 (0.0500)  time: 0.4758  data: 0.0377  max mem: 21002
Epoch: [278]  [130/312]  eta: 0:01:54  lr: 0.000059  min_lr: 0.000059  loss: 3.6597 (3.5801)  weight_decay: 0.0500 (0.0500)  time: 0.7012  data: 0.0518  max mem: 21002
Epoch: [278]  [140/312]  eta: 0:01:43  lr: 0.000059  min_lr: 0.000059  loss: 3.8170 (3.5815)  weight_decay: 0.0500 (0.0500)  time: 0.5238  data: 0.0196  max mem: 21002
Epoch: [278]  [150/312]  eta: 0:01:38  lr: 0.000059  min_lr: 0.000059  loss: 3.5716 (3.5803)  weight_decay: 0.0500 (0.0500)  time: 0.5032  data: 0.0203  max mem: 21002
Epoch: [278]  [160/312]  eta: 0:01:32  lr: 0.000059  min_lr: 0.000059  loss: 3.5200 (3.5733)  weight_decay: 0.0500 (0.0500)  time: 0.6208  data: 0.0673  max mem: 21002
Epoch: [278]  [170/312]  eta: 0:01:23  lr: 0.000059  min_lr: 0.000059  loss: 3.4876 (3.5768)  weight_decay: 0.0500 (0.0500)  time: 0.4165  data: 0.0523  max mem: 21002
Epoch: [278]  [180/312]  eta: 0:01:19  lr: 0.000058  min_lr: 0.000058  loss: 3.4876 (3.5644)  weight_decay: 0.0500 (0.0500)  time: 0.5681  data: 0.0196  max mem: 21002
Epoch: [278]  [190/312]  eta: 0:01:11  lr: 0.000058  min_lr: 0.000058  loss: 3.7357 (3.5704)  weight_decay: 0.0500 (0.0500)  time: 0.5657  data: 0.0194  max mem: 21002
Epoch: [278]  [200/312]  eta: 0:01:05  lr: 0.000058  min_lr: 0.000058  loss: 3.7283 (3.5629)  weight_decay: 0.0500 (0.0500)  time: 0.4114  data: 0.0399  max mem: 21002
Epoch: [278]  [210/312]  eta: 0:01:00  lr: 0.000058  min_lr: 0.000058  loss: 3.6571 (3.5596)  weight_decay: 0.0500 (0.0500)  time: 0.6524  data: 0.0882  max mem: 21002
Epoch: [278]  [220/312]  eta: 0:00:53  lr: 0.000058  min_lr: 0.000058  loss: 3.6571 (3.5642)  weight_decay: 0.0500 (0.0500)  time: 0.5315  data: 0.0491  max mem: 21002
Epoch: [278]  [230/312]  eta: 0:00:47  lr: 0.000058  min_lr: 0.000058  loss: 3.6713 (3.5554)  weight_decay: 0.0500 (0.0500)  time: 0.4695  data: 0.0197  max mem: 21002
Epoch: [278]  [240/312]  eta: 0:00:42  lr: 0.000057  min_lr: 0.000057  loss: 3.6834 (3.5653)  weight_decay: 0.0500 (0.0500)  time: 0.6593  data: 0.1038  max mem: 21002
Epoch: [278]  [250/312]  eta: 0:00:35  lr: 0.000057  min_lr: 0.000057  loss: 3.5519 (3.5542)  weight_decay: 0.0500 (0.0500)  time: 0.4776  data: 0.0849  max mem: 21002
Epoch: [278]  [260/312]  eta: 0:00:30  lr: 0.000057  min_lr: 0.000057  loss: 3.6210 (3.5703)  weight_decay: 0.0500 (0.0500)  time: 0.5220  data: 0.0391  max mem: 21002
Epoch: [278]  [270/312]  eta: 0:00:23  lr: 0.000057  min_lr: 0.000057  loss: 3.9363 (3.5834)  weight_decay: 0.0500 (0.0500)  time: 0.5373  data: 0.0391  max mem: 21002
Epoch: [278]  [280/312]  eta: 0:00:18  lr: 0.000057  min_lr: 0.000057  loss: 3.8221 (3.5906)  weight_decay: 0.0500 (0.0500)  time: 0.5185  data: 0.0473  max mem: 21002
Epoch: [278]  [290/312]  eta: 0:00:12  lr: 0.000057  min_lr: 0.000057  loss: 3.7993 (3.5881)  weight_decay: 0.0500 (0.0500)  time: 0.5987  data: 0.0511  max mem: 21002
Epoch: [278]  [300/312]  eta: 0:00:06  lr: 0.000056  min_lr: 0.000056  loss: 3.6931 (3.5913)  weight_decay: 0.0500 (0.0500)  time: 0.3780  data: 0.0042  max mem: 21002
Epoch: [278]  [310/312]  eta: 0:00:01  lr: 0.000056  min_lr: 0.000056  loss: 3.6888 (3.5922)  weight_decay: 0.0500 (0.0500)  time: 0.2771  data: 0.0001  max mem: 21002
Epoch: [278]  [311/312]  eta: 0:00:00  lr: 0.000056  min_lr: 0.000056  loss: 3.5470 (3.5920)  weight_decay: 0.0500 (0.0500)  time: 0.2770  data: 0.0001  max mem: 21002
Epoch: [278] Total time: 0:02:53 (0.5563 s / it)
Averaged stats: lr: 0.000056  min_lr: 0.000056  loss: 3.5470 (3.5767)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.0507 (1.0507)  acc1: 81.1198 (81.1198)  acc5: 95.7031 (95.7031)  time: 8.4250  data: 8.3063  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3810 (1.3015)  acc1: 73.1771 (73.6960)  acc5: 91.6667 (91.9360)  time: 1.0208  data: 0.9230  max mem: 21002
Test: Total time: 0:00:09 (1.0317 s / it)
* Acc@1 73.650 Acc@5 91.994 loss 1.308
Accuracy of the model on the 50000 test images: 73.7%
Max accuracy: 73.69%
Epoch: [279]  [  0/312]  eta: 1:04:58  lr: 0.000056  min_lr: 0.000056  loss: 3.9999 (3.9999)  weight_decay: 0.0500 (0.0500)  time: 12.4946  data: 10.5417  max mem: 21002
Epoch: [279]  [ 10/312]  eta: 0:08:30  lr: 0.000056  min_lr: 0.000056  loss: 3.7510 (3.8174)  weight_decay: 0.0500 (0.0500)  time: 1.6888  data: 0.9973  max mem: 21002
Epoch: [279]  [ 20/312]  eta: 0:05:17  lr: 0.000056  min_lr: 0.000056  loss: 3.7685 (3.8177)  weight_decay: 0.0500 (0.0500)  time: 0.5178  data: 0.0218  max mem: 21002
Epoch: [279]  [ 30/312]  eta: 0:03:54  lr: 0.000056  min_lr: 0.000056  loss: 3.7913 (3.7010)  weight_decay: 0.0500 (0.0500)  time: 0.3600  data: 0.0010  max mem: 21002
Epoch: [279]  [ 40/312]  eta: 0:03:20  lr: 0.000056  min_lr: 0.000056  loss: 3.4966 (3.6560)  weight_decay: 0.0500 (0.0500)  time: 0.3700  data: 0.0010  max mem: 21002
Epoch: [279]  [ 50/312]  eta: 0:03:09  lr: 0.000055  min_lr: 0.000055  loss: 3.5381 (3.6431)  weight_decay: 0.0500 (0.0500)  time: 0.5507  data: 0.0010  max mem: 21002
Epoch: [279]  [ 60/312]  eta: 0:02:43  lr: 0.000055  min_lr: 0.000055  loss: 3.5447 (3.5930)  weight_decay: 0.0500 (0.0500)  time: 0.4710  data: 0.0010  max mem: 21002
Epoch: [279]  [ 70/312]  eta: 0:02:39  lr: 0.000055  min_lr: 0.000055  loss: 3.7313 (3.6165)  weight_decay: 0.0500 (0.0500)  time: 0.4976  data: 0.0010  max mem: 21002
Epoch: [279]  [ 80/312]  eta: 0:02:31  lr: 0.000055  min_lr: 0.000055  loss: 3.6953 (3.5947)  weight_decay: 0.0500 (0.0500)  time: 0.6637  data: 0.0009  max mem: 21002
Epoch: [279]  [ 90/312]  eta: 0:02:19  lr: 0.000055  min_lr: 0.000055  loss: 3.4984 (3.5927)  weight_decay: 0.0500 (0.0500)  time: 0.5169  data: 0.0410  max mem: 21002
Epoch: [279]  [100/312]  eta: 0:02:13  lr: 0.000055  min_lr: 0.000055  loss: 3.6821 (3.5934)  weight_decay: 0.0500 (0.0500)  time: 0.5395  data: 0.0666  max mem: 21002
Epoch: [279]  [110/312]  eta: 0:02:02  lr: 0.000054  min_lr: 0.000054  loss: 3.8279 (3.5926)  weight_decay: 0.0500 (0.0500)  time: 0.5045  data: 0.0553  max mem: 21002
Epoch: [279]  [120/312]  eta: 0:01:57  lr: 0.000054  min_lr: 0.000054  loss: 3.4009 (3.5651)  weight_decay: 0.0500 (0.0500)  time: 0.5030  data: 0.0886  max mem: 21002
Epoch: [279]  [130/312]  eta: 0:01:50  lr: 0.000054  min_lr: 0.000054  loss: 3.4009 (3.5647)  weight_decay: 0.0500 (0.0500)  time: 0.6357  data: 0.0841  max mem: 21002
Epoch: [279]  [140/312]  eta: 0:01:41  lr: 0.000054  min_lr: 0.000054  loss: 3.7455 (3.5697)  weight_decay: 0.0500 (0.0500)  time: 0.4567  data: 0.0371  max mem: 21002
Epoch: [279]  [150/312]  eta: 0:01:36  lr: 0.000054  min_lr: 0.000054  loss: 3.8068 (3.5843)  weight_decay: 0.0500 (0.0500)  time: 0.5238  data: 0.0649  max mem: 21002
Epoch: [279]  [160/312]  eta: 0:01:29  lr: 0.000054  min_lr: 0.000054  loss: 3.7695 (3.5815)  weight_decay: 0.0500 (0.0500)  time: 0.6061  data: 0.1066  max mem: 21002
Epoch: [279]  [170/312]  eta: 0:01:23  lr: 0.000053  min_lr: 0.000053  loss: 3.7957 (3.6001)  weight_decay: 0.0500 (0.0500)  time: 0.5095  data: 0.1011  max mem: 21002
Epoch: [279]  [180/312]  eta: 0:01:18  lr: 0.000053  min_lr: 0.000053  loss: 3.8022 (3.6079)  weight_decay: 0.0500 (0.0500)  time: 0.6049  data: 0.0743  max mem: 21002
Epoch: [279]  [190/312]  eta: 0:01:10  lr: 0.000053  min_lr: 0.000053  loss: 3.6572 (3.6015)  weight_decay: 0.0500 (0.0500)  time: 0.5097  data: 0.0647  max mem: 21002
Epoch: [279]  [200/312]  eta: 0:01:05  lr: 0.000053  min_lr: 0.000053  loss: 3.3351 (3.5945)  weight_decay: 0.0500 (0.0500)  time: 0.5085  data: 0.0849  max mem: 21002
Epoch: [279]  [210/312]  eta: 0:00:59  lr: 0.000053  min_lr: 0.000053  loss: 3.5092 (3.5890)  weight_decay: 0.0500 (0.0500)  time: 0.6502  data: 0.1034  max mem: 21002
Epoch: [279]  [220/312]  eta: 0:00:53  lr: 0.000053  min_lr: 0.000053  loss: 3.5366 (3.5819)  weight_decay: 0.0500 (0.0500)  time: 0.5111  data: 0.1043  max mem: 21002
Epoch: [279]  [230/312]  eta: 0:00:47  lr: 0.000052  min_lr: 0.000052  loss: 3.6181 (3.5829)  weight_decay: 0.0500 (0.0500)  time: 0.4998  data: 0.0853  max mem: 21002
Epoch: [279]  [240/312]  eta: 0:00:41  lr: 0.000052  min_lr: 0.000052  loss: 3.5554 (3.5813)  weight_decay: 0.0500 (0.0500)  time: 0.5486  data: 0.0895  max mem: 21002
Epoch: [279]  [250/312]  eta: 0:00:35  lr: 0.000052  min_lr: 0.000052  loss: 3.6110 (3.5893)  weight_decay: 0.0500 (0.0500)  time: 0.5002  data: 0.0884  max mem: 21002
Epoch: [279]  [260/312]  eta: 0:00:29  lr: 0.000052  min_lr: 0.000052  loss: 3.6385 (3.5870)  weight_decay: 0.0500 (0.0500)  time: 0.5903  data: 0.0500  max mem: 21002
Epoch: [279]  [270/312]  eta: 0:00:23  lr: 0.000052  min_lr: 0.000052  loss: 3.6336 (3.5801)  weight_decay: 0.0500 (0.0500)  time: 0.4976  data: 0.0403  max mem: 21002
Epoch: [279]  [280/312]  eta: 0:00:18  lr: 0.000052  min_lr: 0.000052  loss: 3.3719 (3.5734)  weight_decay: 0.0500 (0.0500)  time: 0.5204  data: 0.0827  max mem: 21002
Epoch: [279]  [290/312]  eta: 0:00:12  lr: 0.000051  min_lr: 0.000051  loss: 3.2787 (3.5647)  weight_decay: 0.0500 (0.0500)  time: 0.6064  data: 0.0942  max mem: 21002
Epoch: [279]  [300/312]  eta: 0:00:06  lr: 0.000051  min_lr: 0.000051  loss: 3.1839 (3.5566)  weight_decay: 0.0500 (0.0500)  time: 0.4059  data: 0.0500  max mem: 21002
Epoch: [279]  [310/312]  eta: 0:00:01  lr: 0.000051  min_lr: 0.000051  loss: 3.6543 (3.5686)  weight_decay: 0.0500 (0.0500)  time: 0.2890  data: 0.0124  max mem: 21002
Epoch: [279]  [311/312]  eta: 0:00:00  lr: 0.000051  min_lr: 0.000051  loss: 3.8015 (3.5696)  weight_decay: 0.0500 (0.0500)  time: 0.2889  data: 0.0124  max mem: 21002
Epoch: [279] Total time: 0:02:53 (0.5546 s / it)
Averaged stats: lr: 0.000051  min_lr: 0.000051  loss: 3.8015 (3.5665)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 1.0866 (1.0866)  acc1: 81.1198 (81.1198)  acc5: 95.8333 (95.8333)  time: 8.0546  data: 7.9362  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4335 (1.3363)  acc1: 72.7865 (73.9040)  acc5: 91.9271 (92.0480)  time: 1.0004  data: 0.9028  max mem: 21002
Test: Total time: 0:00:09 (1.0100 s / it)
* Acc@1 73.708 Acc@5 92.024 loss 1.341
Accuracy of the model on the 50000 test images: 73.7%
Max accuracy: 73.71%
Epoch: [280]  [  0/312]  eta: 0:58:46  lr: 0.000051  min_lr: 0.000051  loss: 3.5431 (3.5431)  weight_decay: 0.0500 (0.0500)  time: 11.3030  data: 9.8386  max mem: 21002
Epoch: [280]  [ 10/312]  eta: 0:08:29  lr: 0.000051  min_lr: 0.000051  loss: 3.5431 (3.5943)  weight_decay: 0.0500 (0.0500)  time: 1.6887  data: 1.1095  max mem: 21002
Epoch: [280]  [ 20/312]  eta: 0:05:12  lr: 0.000051  min_lr: 0.000051  loss: 3.5238 (3.5312)  weight_decay: 0.0500 (0.0500)  time: 0.5583  data: 0.1226  max mem: 21002
Epoch: [280]  [ 30/312]  eta: 0:03:50  lr: 0.000051  min_lr: 0.000051  loss: 3.6158 (3.5810)  weight_decay: 0.0500 (0.0500)  time: 0.3402  data: 0.0046  max mem: 21002
Epoch: [280]  [ 40/312]  eta: 0:03:14  lr: 0.000050  min_lr: 0.000050  loss: 3.6697 (3.6151)  weight_decay: 0.0500 (0.0500)  time: 0.3401  data: 0.0013  max mem: 21002
Epoch: [280]  [ 50/312]  eta: 0:03:02  lr: 0.000050  min_lr: 0.000050  loss: 3.6726 (3.6370)  weight_decay: 0.0500 (0.0500)  time: 0.5093  data: 0.0285  max mem: 21002
Epoch: [280]  [ 60/312]  eta: 0:02:40  lr: 0.000050  min_lr: 0.000050  loss: 3.8673 (3.6483)  weight_decay: 0.0500 (0.0500)  time: 0.4765  data: 0.0449  max mem: 21002
Epoch: [280]  [ 70/312]  eta: 0:02:35  lr: 0.000050  min_lr: 0.000050  loss: 3.7028 (3.6332)  weight_decay: 0.0500 (0.0500)  time: 0.4999  data: 0.0746  max mem: 21002
Epoch: [280]  [ 80/312]  eta: 0:02:25  lr: 0.000050  min_lr: 0.000050  loss: 3.6632 (3.6351)  weight_decay: 0.0500 (0.0500)  time: 0.6009  data: 0.1029  max mem: 21002
Epoch: [280]  [ 90/312]  eta: 0:02:15  lr: 0.000050  min_lr: 0.000050  loss: 3.5751 (3.6196)  weight_decay: 0.0500 (0.0500)  time: 0.4893  data: 0.0767  max mem: 21002
Epoch: [280]  [100/312]  eta: 0:02:11  lr: 0.000050  min_lr: 0.000050  loss: 3.5016 (3.6085)  weight_decay: 0.0500 (0.0500)  time: 0.6032  data: 0.0827  max mem: 21002
Epoch: [280]  [110/312]  eta: 0:02:00  lr: 0.000049  min_lr: 0.000049  loss: 3.5016 (3.6038)  weight_decay: 0.0500 (0.0500)  time: 0.5406  data: 0.0666  max mem: 21002
Epoch: [280]  [120/312]  eta: 0:01:55  lr: 0.000049  min_lr: 0.000049  loss: 3.4848 (3.5943)  weight_decay: 0.0500 (0.0500)  time: 0.4877  data: 0.0374  max mem: 21002
Epoch: [280]  [130/312]  eta: 0:01:50  lr: 0.000049  min_lr: 0.000049  loss: 3.6746 (3.5955)  weight_decay: 0.0500 (0.0500)  time: 0.6550  data: 0.0872  max mem: 21002
Epoch: [280]  [140/312]  eta: 0:01:40  lr: 0.000049  min_lr: 0.000049  loss: 3.4459 (3.5666)  weight_decay: 0.0500 (0.0500)  time: 0.4811  data: 0.0711  max mem: 21002
Epoch: [280]  [150/312]  eta: 0:01:35  lr: 0.000049  min_lr: 0.000049  loss: 3.3882 (3.5711)  weight_decay: 0.0500 (0.0500)  time: 0.5078  data: 0.0416  max mem: 21002
Epoch: [280]  [160/312]  eta: 0:01:28  lr: 0.000049  min_lr: 0.000049  loss: 3.6710 (3.5762)  weight_decay: 0.0500 (0.0500)  time: 0.5935  data: 0.0504  max mem: 21002
Epoch: [280]  [170/312]  eta: 0:01:22  lr: 0.000048  min_lr: 0.000048  loss: 3.7123 (3.5898)  weight_decay: 0.0500 (0.0500)  time: 0.4781  data: 0.0541  max mem: 21002
Epoch: [280]  [180/312]  eta: 0:01:17  lr: 0.000048  min_lr: 0.000048  loss: 3.7571 (3.5834)  weight_decay: 0.0500 (0.0500)  time: 0.5811  data: 0.0766  max mem: 21002
Epoch: [280]  [190/312]  eta: 0:01:09  lr: 0.000048  min_lr: 0.000048  loss: 3.7767 (3.5877)  weight_decay: 0.0500 (0.0500)  time: 0.5219  data: 0.0786  max mem: 21002
Epoch: [280]  [200/312]  eta: 0:01:04  lr: 0.000048  min_lr: 0.000048  loss: 3.8386 (3.5894)  weight_decay: 0.0500 (0.0500)  time: 0.5398  data: 0.0945  max mem: 21002
Epoch: [280]  [210/312]  eta: 0:00:59  lr: 0.000048  min_lr: 0.000048  loss: 3.7886 (3.5915)  weight_decay: 0.0500 (0.0500)  time: 0.6868  data: 0.1008  max mem: 21002
Epoch: [280]  [220/312]  eta: 0:00:52  lr: 0.000048  min_lr: 0.000048  loss: 3.5654 (3.5882)  weight_decay: 0.0500 (0.0500)  time: 0.4986  data: 0.0665  max mem: 21002
Epoch: [280]  [230/312]  eta: 0:00:47  lr: 0.000048  min_lr: 0.000048  loss: 3.5798 (3.5843)  weight_decay: 0.0500 (0.0500)  time: 0.4928  data: 0.0650  max mem: 21002
Epoch: [280]  [240/312]  eta: 0:00:41  lr: 0.000047  min_lr: 0.000047  loss: 3.6906 (3.5879)  weight_decay: 0.0500 (0.0500)  time: 0.6064  data: 0.0637  max mem: 21002
Epoch: [280]  [250/312]  eta: 0:00:35  lr: 0.000047  min_lr: 0.000047  loss: 3.5602 (3.5761)  weight_decay: 0.0500 (0.0500)  time: 0.5119  data: 0.1009  max mem: 21002
Epoch: [280]  [260/312]  eta: 0:00:29  lr: 0.000047  min_lr: 0.000047  loss: 3.3766 (3.5700)  weight_decay: 0.0500 (0.0500)  time: 0.5538  data: 0.0968  max mem: 21002
Epoch: [280]  [270/312]  eta: 0:00:23  lr: 0.000047  min_lr: 0.000047  loss: 3.6814 (3.5710)  weight_decay: 0.0500 (0.0500)  time: 0.5523  data: 0.0753  max mem: 21002
Epoch: [280]  [280/312]  eta: 0:00:18  lr: 0.000047  min_lr: 0.000047  loss: 3.7234 (3.5627)  weight_decay: 0.0500 (0.0500)  time: 0.4906  data: 0.0659  max mem: 21002
Epoch: [280]  [290/312]  eta: 0:00:12  lr: 0.000047  min_lr: 0.000047  loss: 3.5634 (3.5608)  weight_decay: 0.0500 (0.0500)  time: 0.5664  data: 0.0297  max mem: 21002
Epoch: [280]  [300/312]  eta: 0:00:06  lr: 0.000046  min_lr: 0.000046  loss: 3.6327 (3.5630)  weight_decay: 0.0500 (0.0500)  time: 0.4504  data: 0.0330  max mem: 21002
Epoch: [280]  [310/312]  eta: 0:00:01  lr: 0.000046  min_lr: 0.000046  loss: 3.8362 (3.5695)  weight_decay: 0.0500 (0.0500)  time: 0.2847  data: 0.0078  max mem: 21002
Epoch: [280]  [311/312]  eta: 0:00:00  lr: 0.000046  min_lr: 0.000046  loss: 3.8362 (3.5701)  weight_decay: 0.0500 (0.0500)  time: 0.2845  data: 0.0078  max mem: 21002
Epoch: [280] Total time: 0:02:52 (0.5539 s / it)
Averaged stats: lr: 0.000046  min_lr: 0.000046  loss: 3.8362 (3.5750)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.1049 (1.1049)  acc1: 80.9896 (80.9896)  acc5: 95.8333 (95.8333)  time: 8.6935  data: 8.5751  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4375 (1.3423)  acc1: 72.6562 (73.8400)  acc5: 91.7969 (91.8720)  time: 1.0521  data: 0.9529  max mem: 21002
Test: Total time: 0:00:09 (1.0624 s / it)
* Acc@1 73.586 Acc@5 91.964 loss 1.348
Accuracy of the model on the 50000 test images: 73.6%
Max accuracy: 73.71%
Epoch: [281]  [  0/312]  eta: 0:54:04  lr: 0.000046  min_lr: 0.000046  loss: 4.3696 (4.3696)  weight_decay: 0.0500 (0.0500)  time: 10.3981  data: 9.3807  max mem: 21002
Epoch: [281]  [ 10/312]  eta: 0:08:01  lr: 0.000046  min_lr: 0.000046  loss: 3.9012 (3.6595)  weight_decay: 0.0500 (0.0500)  time: 1.5957  data: 1.0439  max mem: 21002
Epoch: [281]  [ 20/312]  eta: 0:05:20  lr: 0.000046  min_lr: 0.000046  loss: 3.9012 (3.7414)  weight_decay: 0.0500 (0.0500)  time: 0.6330  data: 0.1479  max mem: 21002
Epoch: [281]  [ 30/312]  eta: 0:03:56  lr: 0.000046  min_lr: 0.000046  loss: 3.7914 (3.6112)  weight_decay: 0.0500 (0.0500)  time: 0.4228  data: 0.0433  max mem: 21002
Epoch: [281]  [ 40/312]  eta: 0:03:17  lr: 0.000046  min_lr: 0.000046  loss: 3.7391 (3.6263)  weight_decay: 0.0500 (0.0500)  time: 0.3325  data: 0.0011  max mem: 21002
Epoch: [281]  [ 50/312]  eta: 0:03:06  lr: 0.000046  min_lr: 0.000046  loss: 3.7964 (3.6422)  weight_decay: 0.0500 (0.0500)  time: 0.5148  data: 0.1101  max mem: 21002
Epoch: [281]  [ 60/312]  eta: 0:02:42  lr: 0.000045  min_lr: 0.000045  loss: 3.6622 (3.6296)  weight_decay: 0.0500 (0.0500)  time: 0.4848  data: 0.1229  max mem: 21002
Epoch: [281]  [ 70/312]  eta: 0:02:40  lr: 0.000045  min_lr: 0.000045  loss: 3.7213 (3.6213)  weight_decay: 0.0500 (0.0500)  time: 0.5368  data: 0.1322  max mem: 21002
Epoch: [281]  [ 80/312]  eta: 0:02:28  lr: 0.000045  min_lr: 0.000045  loss: 3.7074 (3.6008)  weight_decay: 0.0500 (0.0500)  time: 0.6270  data: 0.1191  max mem: 21002
Epoch: [281]  [ 90/312]  eta: 0:02:17  lr: 0.000045  min_lr: 0.000045  loss: 3.2943 (3.5803)  weight_decay: 0.0500 (0.0500)  time: 0.4729  data: 0.0769  max mem: 21002
Epoch: [281]  [100/312]  eta: 0:02:12  lr: 0.000045  min_lr: 0.000045  loss: 3.6911 (3.5907)  weight_decay: 0.0500 (0.0500)  time: 0.5668  data: 0.1399  max mem: 21002
Epoch: [281]  [110/312]  eta: 0:02:00  lr: 0.000045  min_lr: 0.000045  loss: 3.7382 (3.6082)  weight_decay: 0.0500 (0.0500)  time: 0.4851  data: 0.0681  max mem: 21002
Epoch: [281]  [120/312]  eta: 0:01:55  lr: 0.000044  min_lr: 0.000044  loss: 3.7382 (3.5994)  weight_decay: 0.0500 (0.0500)  time: 0.4798  data: 0.0767  max mem: 21002
Epoch: [281]  [130/312]  eta: 0:01:50  lr: 0.000044  min_lr: 0.000044  loss: 3.4197 (3.5798)  weight_decay: 0.0500 (0.0500)  time: 0.6820  data: 0.1408  max mem: 21002
Epoch: [281]  [140/312]  eta: 0:01:40  lr: 0.000044  min_lr: 0.000044  loss: 3.5313 (3.5753)  weight_decay: 0.0500 (0.0500)  time: 0.4908  data: 0.0692  max mem: 21002
Epoch: [281]  [150/312]  eta: 0:01:35  lr: 0.000044  min_lr: 0.000044  loss: 3.7480 (3.5964)  weight_decay: 0.0500 (0.0500)  time: 0.4667  data: 0.0676  max mem: 21002
Epoch: [281]  [160/312]  eta: 0:01:29  lr: 0.000044  min_lr: 0.000044  loss: 3.8970 (3.6032)  weight_decay: 0.0500 (0.0500)  time: 0.6068  data: 0.0683  max mem: 21002
Epoch: [281]  [170/312]  eta: 0:01:22  lr: 0.000044  min_lr: 0.000044  loss: 3.8171 (3.6060)  weight_decay: 0.0500 (0.0500)  time: 0.5243  data: 0.0591  max mem: 21002
Epoch: [281]  [180/312]  eta: 0:01:17  lr: 0.000044  min_lr: 0.000044  loss: 3.8281 (3.6164)  weight_decay: 0.0500 (0.0500)  time: 0.5679  data: 0.1164  max mem: 21002
Epoch: [281]  [190/312]  eta: 0:01:09  lr: 0.000043  min_lr: 0.000043  loss: 3.8281 (3.6168)  weight_decay: 0.0500 (0.0500)  time: 0.4898  data: 0.0588  max mem: 21002
Epoch: [281]  [200/312]  eta: 0:01:04  lr: 0.000043  min_lr: 0.000043  loss: 3.8039 (3.6175)  weight_decay: 0.0500 (0.0500)  time: 0.4916  data: 0.0461  max mem: 21002
Epoch: [281]  [210/312]  eta: 0:00:59  lr: 0.000043  min_lr: 0.000043  loss: 3.7203 (3.6156)  weight_decay: 0.0500 (0.0500)  time: 0.6312  data: 0.1018  max mem: 21002
Epoch: [281]  [220/312]  eta: 0:00:52  lr: 0.000043  min_lr: 0.000043  loss: 3.7479 (3.6161)  weight_decay: 0.0500 (0.0500)  time: 0.5084  data: 0.0570  max mem: 21002
Epoch: [281]  [230/312]  eta: 0:00:47  lr: 0.000043  min_lr: 0.000043  loss: 3.6831 (3.6112)  weight_decay: 0.0500 (0.0500)  time: 0.5230  data: 0.0291  max mem: 21002
Epoch: [281]  [240/312]  eta: 0:00:41  lr: 0.000043  min_lr: 0.000043  loss: 3.6595 (3.6165)  weight_decay: 0.0500 (0.0500)  time: 0.5600  data: 0.0286  max mem: 21002
Epoch: [281]  [250/312]  eta: 0:00:35  lr: 0.000043  min_lr: 0.000043  loss: 3.5057 (3.6141)  weight_decay: 0.0500 (0.0500)  time: 0.5317  data: 0.0450  max mem: 21002
Epoch: [281]  [260/312]  eta: 0:00:29  lr: 0.000042  min_lr: 0.000042  loss: 3.5057 (3.6195)  weight_decay: 0.0500 (0.0500)  time: 0.5679  data: 0.0875  max mem: 21002
Epoch: [281]  [270/312]  eta: 0:00:23  lr: 0.000042  min_lr: 0.000042  loss: 3.5310 (3.6062)  weight_decay: 0.0500 (0.0500)  time: 0.5238  data: 0.0440  max mem: 21002
Epoch: [281]  [280/312]  eta: 0:00:18  lr: 0.000042  min_lr: 0.000042  loss: 3.2179 (3.5968)  weight_decay: 0.0500 (0.0500)  time: 0.5971  data: 0.0552  max mem: 21002
Epoch: [281]  [290/312]  eta: 0:00:12  lr: 0.000042  min_lr: 0.000042  loss: 3.3775 (3.5960)  weight_decay: 0.0500 (0.0500)  time: 0.5524  data: 0.0629  max mem: 21002
Epoch: [281]  [300/312]  eta: 0:00:06  lr: 0.000042  min_lr: 0.000042  loss: 3.6684 (3.6052)  weight_decay: 0.0500 (0.0500)  time: 0.3396  data: 0.0089  max mem: 21002
Epoch: [281]  [310/312]  eta: 0:00:01  lr: 0.000042  min_lr: 0.000042  loss: 3.8474 (3.6084)  weight_decay: 0.0500 (0.0500)  time: 0.2774  data: 0.0001  max mem: 21002
Epoch: [281]  [311/312]  eta: 0:00:00  lr: 0.000042  min_lr: 0.000042  loss: 3.8474 (3.6084)  weight_decay: 0.0500 (0.0500)  time: 0.2772  data: 0.0001  max mem: 21002
Epoch: [281] Total time: 0:02:51 (0.5503 s / it)
Averaged stats: lr: 0.000042  min_lr: 0.000042  loss: 3.8474 (3.5969)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.1083 (1.1083)  acc1: 81.2500 (81.2500)  acc5: 95.8333 (95.8333)  time: 8.8682  data: 8.7500  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4293 (1.3371)  acc1: 73.0469 (73.7280)  acc5: 91.1458 (91.9360)  time: 1.0733  data: 0.9723  max mem: 21002
Test: Total time: 0:00:09 (1.1013 s / it)
* Acc@1 73.702 Acc@5 92.020 loss 1.343
Accuracy of the model on the 50000 test images: 73.7%
Max accuracy: 73.71%
Epoch: [282]  [  0/312]  eta: 1:03:27  lr: 0.000042  min_lr: 0.000042  loss: 3.5746 (3.5746)  weight_decay: 0.0500 (0.0500)  time: 12.2039  data: 10.4284  max mem: 21002
Epoch: [282]  [ 10/312]  eta: 0:07:55  lr: 0.000041  min_lr: 0.000041  loss: 3.6864 (3.5272)  weight_decay: 0.0500 (0.0500)  time: 1.5757  data: 1.0435  max mem: 21002
Epoch: [282]  [ 20/312]  eta: 0:05:32  lr: 0.000041  min_lr: 0.000041  loss: 3.6527 (3.4874)  weight_decay: 0.0500 (0.0500)  time: 0.5862  data: 0.1475  max mem: 21002
Epoch: [282]  [ 30/312]  eta: 0:04:04  lr: 0.000041  min_lr: 0.000041  loss: 3.5054 (3.4740)  weight_decay: 0.0500 (0.0500)  time: 0.4752  data: 0.0954  max mem: 21002
Epoch: [282]  [ 40/312]  eta: 0:03:28  lr: 0.000041  min_lr: 0.000041  loss: 3.6432 (3.5322)  weight_decay: 0.0500 (0.0500)  time: 0.3778  data: 0.0291  max mem: 21002
Epoch: [282]  [ 50/312]  eta: 0:03:19  lr: 0.000041  min_lr: 0.000041  loss: 3.7923 (3.5429)  weight_decay: 0.0500 (0.0500)  time: 0.5976  data: 0.1237  max mem: 21002
Epoch: [282]  [ 60/312]  eta: 0:02:52  lr: 0.000041  min_lr: 0.000041  loss: 3.7080 (3.5681)  weight_decay: 0.0500 (0.0500)  time: 0.5087  data: 0.0954  max mem: 21002
Epoch: [282]  [ 70/312]  eta: 0:02:44  lr: 0.000041  min_lr: 0.000041  loss: 3.7080 (3.5634)  weight_decay: 0.0500 (0.0500)  time: 0.4766  data: 0.0941  max mem: 21002
Epoch: [282]  [ 80/312]  eta: 0:02:33  lr: 0.000040  min_lr: 0.000040  loss: 3.6864 (3.5730)  weight_decay: 0.0500 (0.0500)  time: 0.5965  data: 0.1163  max mem: 21002
Epoch: [282]  [ 90/312]  eta: 0:02:21  lr: 0.000040  min_lr: 0.000040  loss: 3.5638 (3.5600)  weight_decay: 0.0500 (0.0500)  time: 0.4781  data: 0.0961  max mem: 21002
Epoch: [282]  [100/312]  eta: 0:02:16  lr: 0.000040  min_lr: 0.000040  loss: 3.6003 (3.5834)  weight_decay: 0.0500 (0.0500)  time: 0.5766  data: 0.1452  max mem: 21002
Epoch: [282]  [110/312]  eta: 0:02:04  lr: 0.000040  min_lr: 0.000040  loss: 3.6095 (3.5770)  weight_decay: 0.0500 (0.0500)  time: 0.5137  data: 0.0800  max mem: 21002
Epoch: [282]  [120/312]  eta: 0:01:58  lr: 0.000040  min_lr: 0.000040  loss: 3.5635 (3.5732)  weight_decay: 0.0500 (0.0500)  time: 0.4687  data: 0.0548  max mem: 21002
Epoch: [282]  [130/312]  eta: 0:01:53  lr: 0.000040  min_lr: 0.000040  loss: 3.6227 (3.5751)  weight_decay: 0.0500 (0.0500)  time: 0.6659  data: 0.1120  max mem: 21002
Epoch: [282]  [140/312]  eta: 0:01:43  lr: 0.000040  min_lr: 0.000040  loss: 3.6910 (3.5583)  weight_decay: 0.0500 (0.0500)  time: 0.5026  data: 0.0761  max mem: 21002
Epoch: [282]  [150/312]  eta: 0:01:38  lr: 0.000040  min_lr: 0.000040  loss: 3.0700 (3.5456)  weight_decay: 0.0500 (0.0500)  time: 0.5224  data: 0.0891  max mem: 21002
Epoch: [282]  [160/312]  eta: 0:01:32  lr: 0.000039  min_lr: 0.000039  loss: 3.5941 (3.5595)  weight_decay: 0.0500 (0.0500)  time: 0.6504  data: 0.0788  max mem: 21002
Epoch: [282]  [170/312]  eta: 0:01:23  lr: 0.000039  min_lr: 0.000039  loss: 3.8284 (3.5667)  weight_decay: 0.0500 (0.0500)  time: 0.4509  data: 0.0257  max mem: 21002
Epoch: [282]  [180/312]  eta: 0:01:19  lr: 0.000039  min_lr: 0.000039  loss: 3.7574 (3.5734)  weight_decay: 0.0500 (0.0500)  time: 0.5599  data: 0.1131  max mem: 21002
Epoch: [282]  [190/312]  eta: 0:01:11  lr: 0.000039  min_lr: 0.000039  loss: 3.6784 (3.5595)  weight_decay: 0.0500 (0.0500)  time: 0.5549  data: 0.1082  max mem: 21002
Epoch: [282]  [200/312]  eta: 0:01:05  lr: 0.000039  min_lr: 0.000039  loss: 3.6774 (3.5625)  weight_decay: 0.0500 (0.0500)  time: 0.4395  data: 0.0314  max mem: 21002
Epoch: [282]  [210/312]  eta: 0:01:00  lr: 0.000039  min_lr: 0.000039  loss: 3.7429 (3.5634)  weight_decay: 0.0500 (0.0500)  time: 0.6465  data: 0.0613  max mem: 21002
Epoch: [282]  [220/312]  eta: 0:00:53  lr: 0.000039  min_lr: 0.000039  loss: 3.7341 (3.5725)  weight_decay: 0.0500 (0.0500)  time: 0.5643  data: 0.0983  max mem: 21002
Epoch: [282]  [230/312]  eta: 0:00:48  lr: 0.000038  min_lr: 0.000038  loss: 3.7341 (3.5741)  weight_decay: 0.0500 (0.0500)  time: 0.5231  data: 0.0723  max mem: 21002
Epoch: [282]  [240/312]  eta: 0:00:42  lr: 0.000038  min_lr: 0.000038  loss: 3.5200 (3.5726)  weight_decay: 0.0500 (0.0500)  time: 0.6653  data: 0.0514  max mem: 21002
Epoch: [282]  [250/312]  eta: 0:00:35  lr: 0.000038  min_lr: 0.000038  loss: 3.3319 (3.5596)  weight_decay: 0.0500 (0.0500)  time: 0.5104  data: 0.0591  max mem: 21002
Epoch: [282]  [260/312]  eta: 0:00:30  lr: 0.000038  min_lr: 0.000038  loss: 3.0955 (3.5484)  weight_decay: 0.0500 (0.0500)  time: 0.5191  data: 0.0640  max mem: 21002
Epoch: [282]  [270/312]  eta: 0:00:24  lr: 0.000038  min_lr: 0.000038  loss: 3.0955 (3.5387)  weight_decay: 0.0500 (0.0500)  time: 0.5112  data: 0.0532  max mem: 21002
Epoch: [282]  [280/312]  eta: 0:00:18  lr: 0.000038  min_lr: 0.000038  loss: 3.1838 (3.5420)  weight_decay: 0.0500 (0.0500)  time: 0.4920  data: 0.0706  max mem: 21002
Epoch: [282]  [290/312]  eta: 0:00:12  lr: 0.000038  min_lr: 0.000038  loss: 3.7427 (3.5445)  weight_decay: 0.0500 (0.0500)  time: 0.5395  data: 0.0821  max mem: 21002
Epoch: [282]  [300/312]  eta: 0:00:06  lr: 0.000037  min_lr: 0.000037  loss: 3.5066 (3.5359)  weight_decay: 0.0500 (0.0500)  time: 0.3533  data: 0.0329  max mem: 21002
Epoch: [282]  [310/312]  eta: 0:00:01  lr: 0.000037  min_lr: 0.000037  loss: 3.3262 (3.5290)  weight_decay: 0.0500 (0.0500)  time: 0.2770  data: 0.0001  max mem: 21002
Epoch: [282]  [311/312]  eta: 0:00:00  lr: 0.000037  min_lr: 0.000037  loss: 3.3262 (3.5306)  weight_decay: 0.0500 (0.0500)  time: 0.2769  data: 0.0001  max mem: 21002
Epoch: [282] Total time: 0:02:53 (0.5566 s / it)
Averaged stats: lr: 0.000037  min_lr: 0.000037  loss: 3.3262 (3.5708)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 0.9875 (0.9875)  acc1: 81.3802 (81.3802)  acc5: 95.9635 (95.9635)  time: 8.4842  data: 8.3652  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3176 (1.2318)  acc1: 73.5677 (74.0480)  acc5: 91.6667 (92.0800)  time: 1.0279  data: 0.9295  max mem: 21002
Test: Total time: 0:00:09 (1.0379 s / it)
* Acc@1 73.986 Acc@5 92.112 loss 1.236
Accuracy of the model on the 50000 test images: 74.0%
Max accuracy: 73.99%
Epoch: [283]  [  0/312]  eta: 1:01:51  lr: 0.000037  min_lr: 0.000037  loss: 3.8518 (3.8518)  weight_decay: 0.0500 (0.0500)  time: 11.8950  data: 11.5969  max mem: 21002
Epoch: [283]  [ 10/312]  eta: 0:08:32  lr: 0.000037  min_lr: 0.000037  loss: 3.6693 (3.5963)  weight_decay: 0.0500 (0.0500)  time: 1.6954  data: 1.0548  max mem: 21002
Epoch: [283]  [ 20/312]  eta: 0:05:03  lr: 0.000037  min_lr: 0.000037  loss: 3.6693 (3.5998)  weight_decay: 0.0500 (0.0500)  time: 0.4954  data: 0.0058  max mem: 21002
Epoch: [283]  [ 30/312]  eta: 0:03:45  lr: 0.000037  min_lr: 0.000037  loss: 3.6971 (3.5850)  weight_decay: 0.0500 (0.0500)  time: 0.3051  data: 0.0062  max mem: 21002
Epoch: [283]  [ 40/312]  eta: 0:03:03  lr: 0.000037  min_lr: 0.000037  loss: 3.6853 (3.5933)  weight_decay: 0.0500 (0.0500)  time: 0.2936  data: 0.0012  max mem: 21002
Epoch: [283]  [ 50/312]  eta: 0:02:53  lr: 0.000037  min_lr: 0.000037  loss: 3.7177 (3.6093)  weight_decay: 0.0500 (0.0500)  time: 0.4521  data: 0.1067  max mem: 21002
Epoch: [283]  [ 60/312]  eta: 0:02:35  lr: 0.000036  min_lr: 0.000036  loss: 3.6680 (3.5782)  weight_decay: 0.0500 (0.0500)  time: 0.4941  data: 0.1243  max mem: 21002
Epoch: [283]  [ 70/312]  eta: 0:02:29  lr: 0.000036  min_lr: 0.000036  loss: 3.5522 (3.5660)  weight_decay: 0.0500 (0.0500)  time: 0.5089  data: 0.1280  max mem: 21002
Epoch: [283]  [ 80/312]  eta: 0:02:25  lr: 0.000036  min_lr: 0.000036  loss: 3.7652 (3.6120)  weight_decay: 0.0500 (0.0500)  time: 0.6577  data: 0.1863  max mem: 21002
Epoch: [283]  [ 90/312]  eta: 0:02:14  lr: 0.000036  min_lr: 0.000036  loss: 3.8056 (3.6174)  weight_decay: 0.0500 (0.0500)  time: 0.5483  data: 0.1017  max mem: 21002
Epoch: [283]  [100/312]  eta: 0:02:08  lr: 0.000036  min_lr: 0.000036  loss: 3.7608 (3.5995)  weight_decay: 0.0500 (0.0500)  time: 0.5213  data: 0.0581  max mem: 21002
Epoch: [283]  [110/312]  eta: 0:01:57  lr: 0.000036  min_lr: 0.000036  loss: 3.1185 (3.5573)  weight_decay: 0.0500 (0.0500)  time: 0.4894  data: 0.0709  max mem: 21002
Epoch: [283]  [120/312]  eta: 0:01:52  lr: 0.000036  min_lr: 0.000036  loss: 3.2324 (3.5603)  weight_decay: 0.0500 (0.0500)  time: 0.4898  data: 0.0724  max mem: 21002
Epoch: [283]  [130/312]  eta: 0:01:47  lr: 0.000036  min_lr: 0.000036  loss: 3.6979 (3.5494)  weight_decay: 0.0500 (0.0500)  time: 0.6292  data: 0.0887  max mem: 21002
Epoch: [283]  [140/312]  eta: 0:01:38  lr: 0.000035  min_lr: 0.000035  loss: 3.6942 (3.5568)  weight_decay: 0.0500 (0.0500)  time: 0.5097  data: 0.0919  max mem: 21002
Epoch: [283]  [150/312]  eta: 0:01:34  lr: 0.000035  min_lr: 0.000035  loss: 3.6942 (3.5649)  weight_decay: 0.0500 (0.0500)  time: 0.5271  data: 0.1007  max mem: 21002
Epoch: [283]  [160/312]  eta: 0:01:27  lr: 0.000035  min_lr: 0.000035  loss: 3.5958 (3.5585)  weight_decay: 0.0500 (0.0500)  time: 0.6016  data: 0.0954  max mem: 21002
Epoch: [283]  [170/312]  eta: 0:01:20  lr: 0.000035  min_lr: 0.000035  loss: 3.6187 (3.5585)  weight_decay: 0.0500 (0.0500)  time: 0.4702  data: 0.0740  max mem: 21002
Epoch: [283]  [180/312]  eta: 0:01:15  lr: 0.000035  min_lr: 0.000035  loss: 3.7579 (3.5692)  weight_decay: 0.0500 (0.0500)  time: 0.5489  data: 0.1152  max mem: 21002
Epoch: [283]  [190/312]  eta: 0:01:08  lr: 0.000035  min_lr: 0.000035  loss: 3.8786 (3.5779)  weight_decay: 0.0500 (0.0500)  time: 0.5126  data: 0.0739  max mem: 21002
Epoch: [283]  [200/312]  eta: 0:01:03  lr: 0.000035  min_lr: 0.000035  loss: 3.8148 (3.5845)  weight_decay: 0.0500 (0.0500)  time: 0.4929  data: 0.0489  max mem: 21002
Epoch: [283]  [210/312]  eta: 0:00:58  lr: 0.000034  min_lr: 0.000034  loss: 3.7090 (3.5839)  weight_decay: 0.0500 (0.0500)  time: 0.6479  data: 0.0871  max mem: 21002
Epoch: [283]  [220/312]  eta: 0:00:51  lr: 0.000034  min_lr: 0.000034  loss: 3.7090 (3.5929)  weight_decay: 0.0500 (0.0500)  time: 0.5151  data: 0.0493  max mem: 21002
Epoch: [283]  [230/312]  eta: 0:00:46  lr: 0.000034  min_lr: 0.000034  loss: 3.7768 (3.5984)  weight_decay: 0.0500 (0.0500)  time: 0.4981  data: 0.0743  max mem: 21002
Epoch: [283]  [240/312]  eta: 0:00:40  lr: 0.000034  min_lr: 0.000034  loss: 3.8320 (3.6168)  weight_decay: 0.0500 (0.0500)  time: 0.5967  data: 0.0649  max mem: 21002
Epoch: [283]  [250/312]  eta: 0:00:34  lr: 0.000034  min_lr: 0.000034  loss: 3.8935 (3.6261)  weight_decay: 0.0500 (0.0500)  time: 0.5389  data: 0.0697  max mem: 21002
Epoch: [283]  [260/312]  eta: 0:00:29  lr: 0.000034  min_lr: 0.000034  loss: 3.8741 (3.6229)  weight_decay: 0.0500 (0.0500)  time: 0.5480  data: 0.1165  max mem: 21002
Epoch: [283]  [270/312]  eta: 0:00:23  lr: 0.000034  min_lr: 0.000034  loss: 3.6531 (3.6211)  weight_decay: 0.0500 (0.0500)  time: 0.4950  data: 0.0539  max mem: 21002
Epoch: [283]  [280/312]  eta: 0:00:17  lr: 0.000034  min_lr: 0.000034  loss: 3.6531 (3.6226)  weight_decay: 0.0500 (0.0500)  time: 0.4917  data: 0.0419  max mem: 21002
Epoch: [283]  [290/312]  eta: 0:00:12  lr: 0.000033  min_lr: 0.000033  loss: 3.8954 (3.6335)  weight_decay: 0.0500 (0.0500)  time: 0.5933  data: 0.0567  max mem: 21002
Epoch: [283]  [300/312]  eta: 0:00:06  lr: 0.000033  min_lr: 0.000033  loss: 3.7894 (3.6348)  weight_decay: 0.0500 (0.0500)  time: 0.4811  data: 0.0209  max mem: 21002
Epoch: [283]  [310/312]  eta: 0:00:01  lr: 0.000033  min_lr: 0.000033  loss: 3.6824 (3.6303)  weight_decay: 0.0500 (0.0500)  time: 0.3225  data: 0.0001  max mem: 21002
Epoch: [283]  [311/312]  eta: 0:00:00  lr: 0.000033  min_lr: 0.000033  loss: 3.7527 (3.6308)  weight_decay: 0.0500 (0.0500)  time: 0.3222  data: 0.0001  max mem: 21002
Epoch: [283] Total time: 0:02:50 (0.5479 s / it)
Averaged stats: lr: 0.000033  min_lr: 0.000033  loss: 3.7527 (3.5814)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 1.0585 (1.0585)  acc1: 81.3802 (81.3802)  acc5: 96.0938 (96.0938)  time: 8.1594  data: 8.0431  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3832 (1.3002)  acc1: 73.3073 (74.0800)  acc5: 92.0573 (92.2560)  time: 1.0085  data: 0.9095  max mem: 21002
Test: Total time: 0:00:09 (1.0193 s / it)
* Acc@1 73.912 Acc@5 92.152 loss 1.305
Accuracy of the model on the 50000 test images: 73.9%
Max accuracy: 73.99%
Epoch: [284]  [  0/312]  eta: 0:57:15  lr: 0.000033  min_lr: 0.000033  loss: 3.6591 (3.6591)  weight_decay: 0.0500 (0.0500)  time: 11.0098  data: 9.0775  max mem: 21002
Epoch: [284]  [ 10/312]  eta: 0:07:48  lr: 0.000033  min_lr: 0.000033  loss: 3.5778 (3.4120)  weight_decay: 0.0500 (0.0500)  time: 1.5524  data: 1.1155  max mem: 21002
Epoch: [284]  [ 20/312]  eta: 0:05:11  lr: 0.000033  min_lr: 0.000033  loss: 3.6195 (3.5190)  weight_decay: 0.0500 (0.0500)  time: 0.5699  data: 0.2374  max mem: 21002
Epoch: [284]  [ 30/312]  eta: 0:03:50  lr: 0.000033  min_lr: 0.000033  loss: 3.7316 (3.5806)  weight_decay: 0.0500 (0.0500)  time: 0.4116  data: 0.0791  max mem: 21002
Epoch: [284]  [ 40/312]  eta: 0:03:28  lr: 0.000033  min_lr: 0.000033  loss: 3.7316 (3.5757)  weight_decay: 0.0500 (0.0500)  time: 0.4518  data: 0.0778  max mem: 21002
Epoch: [284]  [ 50/312]  eta: 0:03:18  lr: 0.000032  min_lr: 0.000032  loss: 3.7756 (3.5908)  weight_decay: 0.0500 (0.0500)  time: 0.6665  data: 0.1894  max mem: 21002
Epoch: [284]  [ 60/312]  eta: 0:02:51  lr: 0.000032  min_lr: 0.000032  loss: 3.7756 (3.5791)  weight_decay: 0.0500 (0.0500)  time: 0.5043  data: 0.1134  max mem: 21002
Epoch: [284]  [ 70/312]  eta: 0:02:43  lr: 0.000032  min_lr: 0.000032  loss: 3.6401 (3.5496)  weight_decay: 0.0500 (0.0500)  time: 0.4660  data: 0.0699  max mem: 21002
Epoch: [284]  [ 80/312]  eta: 0:02:28  lr: 0.000032  min_lr: 0.000032  loss: 3.4464 (3.5397)  weight_decay: 0.0500 (0.0500)  time: 0.5149  data: 0.0697  max mem: 21002
Epoch: [284]  [ 90/312]  eta: 0:02:22  lr: 0.000032  min_lr: 0.000032  loss: 3.6905 (3.5541)  weight_decay: 0.0500 (0.0500)  time: 0.5156  data: 0.1042  max mem: 21002
Epoch: [284]  [100/312]  eta: 0:02:17  lr: 0.000032  min_lr: 0.000032  loss: 3.6459 (3.5447)  weight_decay: 0.0500 (0.0500)  time: 0.6734  data: 0.1801  max mem: 21002
Epoch: [284]  [110/312]  eta: 0:02:04  lr: 0.000032  min_lr: 0.000032  loss: 3.5967 (3.5386)  weight_decay: 0.0500 (0.0500)  time: 0.4961  data: 0.0765  max mem: 21002
Epoch: [284]  [120/312]  eta: 0:01:59  lr: 0.000032  min_lr: 0.000032  loss: 3.3890 (3.5242)  weight_decay: 0.0500 (0.0500)  time: 0.4875  data: 0.0665  max mem: 21002
Epoch: [284]  [130/312]  eta: 0:01:53  lr: 0.000031  min_lr: 0.000031  loss: 3.5455 (3.5375)  weight_decay: 0.0500 (0.0500)  time: 0.6865  data: 0.1044  max mem: 21002
Epoch: [284]  [140/312]  eta: 0:01:43  lr: 0.000031  min_lr: 0.000031  loss: 3.5886 (3.5374)  weight_decay: 0.0500 (0.0500)  time: 0.4921  data: 0.0390  max mem: 21002
Epoch: [284]  [150/312]  eta: 0:01:36  lr: 0.000031  min_lr: 0.000031  loss: 3.5554 (3.5452)  weight_decay: 0.0500 (0.0500)  time: 0.4206  data: 0.0083  max mem: 21002
Epoch: [284]  [160/312]  eta: 0:01:30  lr: 0.000031  min_lr: 0.000031  loss: 3.8431 (3.5752)  weight_decay: 0.0500 (0.0500)  time: 0.5458  data: 0.0080  max mem: 21002
Epoch: [284]  [170/312]  eta: 0:01:23  lr: 0.000031  min_lr: 0.000031  loss: 3.9767 (3.5821)  weight_decay: 0.0500 (0.0500)  time: 0.4832  data: 0.0641  max mem: 21002
Epoch: [284]  [180/312]  eta: 0:01:18  lr: 0.000031  min_lr: 0.000031  loss: 3.8747 (3.5893)  weight_decay: 0.0500 (0.0500)  time: 0.5773  data: 0.1469  max mem: 21002
Epoch: [284]  [190/312]  eta: 0:01:10  lr: 0.000031  min_lr: 0.000031  loss: 3.6416 (3.5862)  weight_decay: 0.0500 (0.0500)  time: 0.5107  data: 0.0837  max mem: 21002
Epoch: [284]  [200/312]  eta: 0:01:05  lr: 0.000031  min_lr: 0.000031  loss: 3.6416 (3.5963)  weight_decay: 0.0500 (0.0500)  time: 0.4905  data: 0.0401  max mem: 21002
Epoch: [284]  [210/312]  eta: 0:00:59  lr: 0.000030  min_lr: 0.000030  loss: 3.7678 (3.6043)  weight_decay: 0.0500 (0.0500)  time: 0.6526  data: 0.0472  max mem: 21002
Epoch: [284]  [220/312]  eta: 0:00:52  lr: 0.000030  min_lr: 0.000030  loss: 3.7349 (3.6140)  weight_decay: 0.0500 (0.0500)  time: 0.4666  data: 0.0246  max mem: 21002
Epoch: [284]  [230/312]  eta: 0:00:47  lr: 0.000030  min_lr: 0.000030  loss: 3.7125 (3.6137)  weight_decay: 0.0500 (0.0500)  time: 0.4904  data: 0.0290  max mem: 21002
Epoch: [284]  [240/312]  eta: 0:00:41  lr: 0.000030  min_lr: 0.000030  loss: 3.7165 (3.6225)  weight_decay: 0.0500 (0.0500)  time: 0.5898  data: 0.0274  max mem: 21002
Epoch: [284]  [250/312]  eta: 0:00:35  lr: 0.000030  min_lr: 0.000030  loss: 3.8765 (3.6293)  weight_decay: 0.0500 (0.0500)  time: 0.5117  data: 0.0447  max mem: 21002
Epoch: [284]  [260/312]  eta: 0:00:29  lr: 0.000030  min_lr: 0.000030  loss: 3.7874 (3.6280)  weight_decay: 0.0500 (0.0500)  time: 0.5498  data: 0.0510  max mem: 21002
Epoch: [284]  [270/312]  eta: 0:00:23  lr: 0.000030  min_lr: 0.000030  loss: 3.7671 (3.6302)  weight_decay: 0.0500 (0.0500)  time: 0.4438  data: 0.0220  max mem: 21002
Epoch: [284]  [280/312]  eta: 0:00:18  lr: 0.000030  min_lr: 0.000030  loss: 3.6828 (3.6203)  weight_decay: 0.0500 (0.0500)  time: 0.5771  data: 0.1186  max mem: 21002
Epoch: [284]  [290/312]  eta: 0:00:12  lr: 0.000030  min_lr: 0.000030  loss: 3.4660 (3.6159)  weight_decay: 0.0500 (0.0500)  time: 0.6446  data: 0.1184  max mem: 21002
Epoch: [284]  [300/312]  eta: 0:00:06  lr: 0.000029  min_lr: 0.000029  loss: 3.4660 (3.6099)  weight_decay: 0.0500 (0.0500)  time: 0.3619  data: 0.0113  max mem: 21002
Epoch: [284]  [310/312]  eta: 0:00:01  lr: 0.000029  min_lr: 0.000029  loss: 3.4282 (3.6052)  weight_decay: 0.0500 (0.0500)  time: 0.2877  data: 0.0113  max mem: 21002
Epoch: [284]  [311/312]  eta: 0:00:00  lr: 0.000029  min_lr: 0.000029  loss: 3.4282 (3.6061)  weight_decay: 0.0500 (0.0500)  time: 0.2770  data: 0.0001  max mem: 21002
Epoch: [284] Total time: 0:02:52 (0.5522 s / it)
Averaged stats: lr: 0.000029  min_lr: 0.000029  loss: 3.4282 (3.5721)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.0166 (1.0166)  acc1: 81.2500 (81.2500)  acc5: 96.0938 (96.0938)  time: 8.6790  data: 8.5605  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3536 (1.2649)  acc1: 73.1771 (74.1440)  acc5: 91.6667 (92.2240)  time: 1.0533  data: 0.9567  max mem: 21002
Test: Total time: 0:00:09 (1.0634 s / it)
* Acc@1 73.890 Acc@5 92.200 loss 1.269
Accuracy of the model on the 50000 test images: 73.9%
Max accuracy: 73.99%
Epoch: [285]  [  0/312]  eta: 1:01:04  lr: 0.000029  min_lr: 0.000029  loss: 2.7616 (2.7616)  weight_decay: 0.0500 (0.0500)  time: 11.7468  data: 8.4125  max mem: 21002
Epoch: [285]  [ 10/312]  eta: 0:08:24  lr: 0.000029  min_lr: 0.000029  loss: 3.6388 (3.3763)  weight_decay: 0.0500 (0.0500)  time: 1.6698  data: 1.0022  max mem: 21002
Epoch: [285]  [ 20/312]  eta: 0:05:13  lr: 0.000029  min_lr: 0.000029  loss: 3.6388 (3.4596)  weight_decay: 0.0500 (0.0500)  time: 0.5392  data: 0.1573  max mem: 21002
Epoch: [285]  [ 30/312]  eta: 0:03:51  lr: 0.000029  min_lr: 0.000029  loss: 3.6805 (3.5212)  weight_decay: 0.0500 (0.0500)  time: 0.3536  data: 0.0270  max mem: 21002
Epoch: [285]  [ 40/312]  eta: 0:03:11  lr: 0.000029  min_lr: 0.000029  loss: 3.7890 (3.5654)  weight_decay: 0.0500 (0.0500)  time: 0.3200  data: 0.0009  max mem: 21002
Epoch: [285]  [ 50/312]  eta: 0:03:04  lr: 0.000029  min_lr: 0.000029  loss: 3.7996 (3.5845)  weight_decay: 0.0500 (0.0500)  time: 0.5230  data: 0.0581  max mem: 21002
Epoch: [285]  [ 60/312]  eta: 0:02:40  lr: 0.000029  min_lr: 0.000029  loss: 3.8338 (3.5700)  weight_decay: 0.0500 (0.0500)  time: 0.4945  data: 0.0579  max mem: 21002
Epoch: [285]  [ 70/312]  eta: 0:02:34  lr: 0.000028  min_lr: 0.000028  loss: 3.7732 (3.5741)  weight_decay: 0.0500 (0.0500)  time: 0.4693  data: 0.0493  max mem: 21002
Epoch: [285]  [ 80/312]  eta: 0:02:26  lr: 0.000028  min_lr: 0.000028  loss: 3.6407 (3.5952)  weight_decay: 0.0500 (0.0500)  time: 0.6141  data: 0.0492  max mem: 21002
Epoch: [285]  [ 90/312]  eta: 0:02:13  lr: 0.000028  min_lr: 0.000028  loss: 3.6407 (3.5891)  weight_decay: 0.0500 (0.0500)  time: 0.4819  data: 0.0365  max mem: 21002
Epoch: [285]  [100/312]  eta: 0:02:10  lr: 0.000028  min_lr: 0.000028  loss: 3.3077 (3.5497)  weight_decay: 0.0500 (0.0500)  time: 0.5524  data: 0.0536  max mem: 21002
Epoch: [285]  [110/312]  eta: 0:01:58  lr: 0.000028  min_lr: 0.000028  loss: 3.4448 (3.5500)  weight_decay: 0.0500 (0.0500)  time: 0.5052  data: 0.0178  max mem: 21002
Epoch: [285]  [120/312]  eta: 0:01:54  lr: 0.000028  min_lr: 0.000028  loss: 3.4094 (3.5230)  weight_decay: 0.0500 (0.0500)  time: 0.4898  data: 0.0204  max mem: 21002
Epoch: [285]  [130/312]  eta: 0:01:49  lr: 0.000028  min_lr: 0.000028  loss: 3.5210 (3.5357)  weight_decay: 0.0500 (0.0500)  time: 0.7000  data: 0.0275  max mem: 21002
Epoch: [285]  [140/312]  eta: 0:01:40  lr: 0.000028  min_lr: 0.000028  loss: 3.8847 (3.5623)  weight_decay: 0.0500 (0.0500)  time: 0.5153  data: 0.0267  max mem: 21002
Epoch: [285]  [150/312]  eta: 0:01:34  lr: 0.000027  min_lr: 0.000027  loss: 3.6164 (3.5510)  weight_decay: 0.0500 (0.0500)  time: 0.4637  data: 0.0375  max mem: 21002
Epoch: [285]  [160/312]  eta: 0:01:28  lr: 0.000027  min_lr: 0.000027  loss: 3.5783 (3.5575)  weight_decay: 0.0500 (0.0500)  time: 0.5618  data: 0.0188  max mem: 21002
Epoch: [285]  [170/312]  eta: 0:01:21  lr: 0.000027  min_lr: 0.000027  loss: 3.7362 (3.5598)  weight_decay: 0.0500 (0.0500)  time: 0.4815  data: 0.0326  max mem: 21002
Epoch: [285]  [180/312]  eta: 0:01:16  lr: 0.000027  min_lr: 0.000027  loss: 3.6072 (3.5566)  weight_decay: 0.0500 (0.0500)  time: 0.5679  data: 0.0637  max mem: 21002
Epoch: [285]  [190/312]  eta: 0:01:09  lr: 0.000027  min_lr: 0.000027  loss: 3.8942 (3.5759)  weight_decay: 0.0500 (0.0500)  time: 0.5089  data: 0.0522  max mem: 21002
Epoch: [285]  [200/312]  eta: 0:01:04  lr: 0.000027  min_lr: 0.000027  loss: 3.7886 (3.5659)  weight_decay: 0.0500 (0.0500)  time: 0.5015  data: 0.0452  max mem: 21002
Epoch: [285]  [210/312]  eta: 0:00:58  lr: 0.000027  min_lr: 0.000027  loss: 3.5396 (3.5656)  weight_decay: 0.0500 (0.0500)  time: 0.6471  data: 0.0301  max mem: 21002
Epoch: [285]  [220/312]  eta: 0:00:52  lr: 0.000027  min_lr: 0.000027  loss: 3.5396 (3.5666)  weight_decay: 0.0500 (0.0500)  time: 0.5082  data: 0.0316  max mem: 21002
Epoch: [285]  [230/312]  eta: 0:00:46  lr: 0.000027  min_lr: 0.000027  loss: 3.8140 (3.5772)  weight_decay: 0.0500 (0.0500)  time: 0.5264  data: 0.0665  max mem: 21002
Epoch: [285]  [240/312]  eta: 0:00:40  lr: 0.000026  min_lr: 0.000026  loss: 3.9079 (3.5868)  weight_decay: 0.0500 (0.0500)  time: 0.5281  data: 0.0688  max mem: 21002
Epoch: [285]  [250/312]  eta: 0:00:35  lr: 0.000026  min_lr: 0.000026  loss: 3.8102 (3.5853)  weight_decay: 0.0500 (0.0500)  time: 0.5295  data: 0.0846  max mem: 21002
Epoch: [285]  [260/312]  eta: 0:00:29  lr: 0.000026  min_lr: 0.000026  loss: 3.6828 (3.5871)  weight_decay: 0.0500 (0.0500)  time: 0.6359  data: 0.0697  max mem: 21002
Epoch: [285]  [270/312]  eta: 0:00:23  lr: 0.000026  min_lr: 0.000026  loss: 3.6828 (3.5834)  weight_decay: 0.0500 (0.0500)  time: 0.4915  data: 0.0496  max mem: 21002
Epoch: [285]  [280/312]  eta: 0:00:18  lr: 0.000026  min_lr: 0.000026  loss: 3.7346 (3.5885)  weight_decay: 0.0500 (0.0500)  time: 0.5349  data: 0.0807  max mem: 21002
Epoch: [285]  [290/312]  eta: 0:00:12  lr: 0.000026  min_lr: 0.000026  loss: 3.7346 (3.5867)  weight_decay: 0.0500 (0.0500)  time: 0.5869  data: 0.0514  max mem: 21002
Epoch: [285]  [300/312]  eta: 0:00:06  lr: 0.000026  min_lr: 0.000026  loss: 3.7708 (3.5896)  weight_decay: 0.0500 (0.0500)  time: 0.4025  data: 0.0194  max mem: 21002
Epoch: [285]  [310/312]  eta: 0:00:01  lr: 0.000026  min_lr: 0.000026  loss: 3.7624 (3.5874)  weight_decay: 0.0500 (0.0500)  time: 0.3050  data: 0.0127  max mem: 21002
Epoch: [285]  [311/312]  eta: 0:00:00  lr: 0.000026  min_lr: 0.000026  loss: 3.6587 (3.5847)  weight_decay: 0.0500 (0.0500)  time: 0.2887  data: 0.0126  max mem: 21002
Epoch: [285] Total time: 0:02:51 (0.5492 s / it)
Averaged stats: lr: 0.000026  min_lr: 0.000026  loss: 3.6587 (3.5750)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.0234 (1.0234)  acc1: 81.5104 (81.5104)  acc5: 95.9635 (95.9635)  time: 8.4013  data: 8.2823  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3445 (1.2597)  acc1: 73.5677 (74.1280)  acc5: 92.0573 (92.0480)  time: 1.0325  data: 0.9336  max mem: 21002
Test: Total time: 0:00:09 (1.0428 s / it)
* Acc@1 73.882 Acc@5 92.176 loss 1.265
Accuracy of the model on the 50000 test images: 73.9%
Max accuracy: 73.99%
Epoch: [286]  [  0/312]  eta: 1:00:13  lr: 0.000026  min_lr: 0.000026  loss: 3.3895 (3.3895)  weight_decay: 0.0500 (0.0500)  time: 11.5825  data: 9.1283  max mem: 21002
Epoch: [286]  [ 10/312]  eta: 0:07:56  lr: 0.000026  min_lr: 0.000026  loss: 3.6815 (3.7661)  weight_decay: 0.0500 (0.0500)  time: 1.5782  data: 0.9148  max mem: 21002
Epoch: [286]  [ 20/312]  eta: 0:05:27  lr: 0.000025  min_lr: 0.000025  loss: 3.6815 (3.6642)  weight_decay: 0.0500 (0.0500)  time: 0.5995  data: 0.1342  max mem: 21002
Epoch: [286]  [ 30/312]  eta: 0:04:01  lr: 0.000025  min_lr: 0.000025  loss: 3.6656 (3.5815)  weight_decay: 0.0500 (0.0500)  time: 0.4590  data: 0.0879  max mem: 21002
Epoch: [286]  [ 40/312]  eta: 0:03:18  lr: 0.000025  min_lr: 0.000025  loss: 3.0310 (3.5120)  weight_decay: 0.0500 (0.0500)  time: 0.3160  data: 0.0007  max mem: 21002
Epoch: [286]  [ 50/312]  eta: 0:03:05  lr: 0.000025  min_lr: 0.000025  loss: 3.5871 (3.5055)  weight_decay: 0.0500 (0.0500)  time: 0.4803  data: 0.1372  max mem: 21002
Epoch: [286]  [ 60/312]  eta: 0:02:41  lr: 0.000025  min_lr: 0.000025  loss: 3.6840 (3.5403)  weight_decay: 0.0500 (0.0500)  time: 0.4585  data: 0.1380  max mem: 21002
Epoch: [286]  [ 70/312]  eta: 0:02:36  lr: 0.000025  min_lr: 0.000025  loss: 3.9168 (3.5860)  weight_decay: 0.0500 (0.0500)  time: 0.4893  data: 0.1313  max mem: 21002
Epoch: [286]  [ 80/312]  eta: 0:02:30  lr: 0.000025  min_lr: 0.000025  loss: 3.9474 (3.6119)  weight_decay: 0.0500 (0.0500)  time: 0.6819  data: 0.2492  max mem: 21002
Epoch: [286]  [ 90/312]  eta: 0:02:15  lr: 0.000025  min_lr: 0.000025  loss: 3.6456 (3.6018)  weight_decay: 0.0500 (0.0500)  time: 0.4811  data: 0.1193  max mem: 21002
Epoch: [286]  [100/312]  eta: 0:02:11  lr: 0.000025  min_lr: 0.000025  loss: 3.3536 (3.5737)  weight_decay: 0.0500 (0.0500)  time: 0.5049  data: 0.1064  max mem: 21002
Epoch: [286]  [110/312]  eta: 0:01:59  lr: 0.000024  min_lr: 0.000024  loss: 3.3161 (3.5658)  weight_decay: 0.0500 (0.0500)  time: 0.5053  data: 0.1063  max mem: 21002
Epoch: [286]  [120/312]  eta: 0:01:55  lr: 0.000024  min_lr: 0.000024  loss: 3.4158 (3.5568)  weight_decay: 0.0500 (0.0500)  time: 0.5026  data: 0.1161  max mem: 21002
Epoch: [286]  [130/312]  eta: 0:01:51  lr: 0.000024  min_lr: 0.000024  loss: 3.6753 (3.5697)  weight_decay: 0.0500 (0.0500)  time: 0.7399  data: 0.2044  max mem: 21002
Epoch: [286]  [140/312]  eta: 0:01:41  lr: 0.000024  min_lr: 0.000024  loss: 3.9095 (3.5862)  weight_decay: 0.0500 (0.0500)  time: 0.5249  data: 0.0898  max mem: 21002
Epoch: [286]  [150/312]  eta: 0:01:35  lr: 0.000024  min_lr: 0.000024  loss: 3.8578 (3.5748)  weight_decay: 0.0500 (0.0500)  time: 0.4478  data: 0.0606  max mem: 21002
Epoch: [286]  [160/312]  eta: 0:01:30  lr: 0.000024  min_lr: 0.000024  loss: 3.5642 (3.5775)  weight_decay: 0.0500 (0.0500)  time: 0.6374  data: 0.1329  max mem: 21002
Epoch: [286]  [170/312]  eta: 0:01:22  lr: 0.000024  min_lr: 0.000024  loss: 3.7960 (3.5942)  weight_decay: 0.0500 (0.0500)  time: 0.5231  data: 0.1198  max mem: 21002
Epoch: [286]  [180/312]  eta: 0:01:17  lr: 0.000024  min_lr: 0.000024  loss: 3.6721 (3.5851)  weight_decay: 0.0500 (0.0500)  time: 0.5208  data: 0.1058  max mem: 21002
Epoch: [286]  [190/312]  eta: 0:01:09  lr: 0.000024  min_lr: 0.000024  loss: 3.5317 (3.5956)  weight_decay: 0.0500 (0.0500)  time: 0.4769  data: 0.0597  max mem: 21002
Epoch: [286]  [200/312]  eta: 0:01:04  lr: 0.000023  min_lr: 0.000023  loss: 3.7619 (3.5963)  weight_decay: 0.0500 (0.0500)  time: 0.4732  data: 0.0657  max mem: 21002
Epoch: [286]  [210/312]  eta: 0:00:59  lr: 0.000023  min_lr: 0.000023  loss: 3.6684 (3.5956)  weight_decay: 0.0500 (0.0500)  time: 0.6961  data: 0.1471  max mem: 21002
Epoch: [286]  [220/312]  eta: 0:00:52  lr: 0.000023  min_lr: 0.000023  loss: 3.6682 (3.5903)  weight_decay: 0.0500 (0.0500)  time: 0.5228  data: 0.0822  max mem: 21002
Epoch: [286]  [230/312]  eta: 0:00:47  lr: 0.000023  min_lr: 0.000023  loss: 3.5151 (3.5829)  weight_decay: 0.0500 (0.0500)  time: 0.4862  data: 0.0608  max mem: 21002
Epoch: [286]  [240/312]  eta: 0:00:41  lr: 0.000023  min_lr: 0.000023  loss: 3.5151 (3.5869)  weight_decay: 0.0500 (0.0500)  time: 0.6360  data: 0.0790  max mem: 21002
Epoch: [286]  [250/312]  eta: 0:00:35  lr: 0.000023  min_lr: 0.000023  loss: 3.8179 (3.5912)  weight_decay: 0.0500 (0.0500)  time: 0.4949  data: 0.0456  max mem: 21002
Epoch: [286]  [260/312]  eta: 0:00:29  lr: 0.000023  min_lr: 0.000023  loss: 3.8712 (3.5985)  weight_decay: 0.0500 (0.0500)  time: 0.5439  data: 0.0798  max mem: 21002
Epoch: [286]  [270/312]  eta: 0:00:23  lr: 0.000023  min_lr: 0.000023  loss: 3.8630 (3.5988)  weight_decay: 0.0500 (0.0500)  time: 0.5021  data: 0.0536  max mem: 21002
Epoch: [286]  [280/312]  eta: 0:00:18  lr: 0.000023  min_lr: 0.000023  loss: 3.7781 (3.6038)  weight_decay: 0.0500 (0.0500)  time: 0.4858  data: 0.0322  max mem: 21002
Epoch: [286]  [290/312]  eta: 0:00:12  lr: 0.000022  min_lr: 0.000022  loss: 3.6998 (3.6045)  weight_decay: 0.0500 (0.0500)  time: 0.5904  data: 0.0518  max mem: 21002
Epoch: [286]  [300/312]  eta: 0:00:06  lr: 0.000022  min_lr: 0.000022  loss: 3.6220 (3.6022)  weight_decay: 0.0500 (0.0500)  time: 0.3915  data: 0.0205  max mem: 21002
Epoch: [286]  [310/312]  eta: 0:00:01  lr: 0.000022  min_lr: 0.000022  loss: 3.6439 (3.6019)  weight_decay: 0.0500 (0.0500)  time: 0.2771  data: 0.0001  max mem: 21002
Epoch: [286]  [311/312]  eta: 0:00:00  lr: 0.000022  min_lr: 0.000022  loss: 3.6220 (3.6010)  weight_decay: 0.0500 (0.0500)  time: 0.2770  data: 0.0001  max mem: 21002
Epoch: [286] Total time: 0:02:51 (0.5506 s / it)
Averaged stats: lr: 0.000022  min_lr: 0.000022  loss: 3.6220 (3.5739)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.0496 (1.0496)  acc1: 81.3802 (81.3802)  acc5: 95.9635 (95.9635)  time: 8.5020  data: 8.3832  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3748 (1.2870)  acc1: 73.3073 (73.8720)  acc5: 91.6667 (92.1760)  time: 1.0399  data: 0.9315  max mem: 21002
Test: Total time: 0:00:09 (1.0511 s / it)
* Acc@1 73.842 Acc@5 92.160 loss 1.292
Accuracy of the model on the 50000 test images: 73.8%
Max accuracy: 73.99%
Epoch: [287]  [  0/312]  eta: 1:03:50  lr: 0.000022  min_lr: 0.000022  loss: 4.2577 (4.2577)  weight_decay: 0.0500 (0.0500)  time: 12.2780  data: 9.0011  max mem: 21002
Epoch: [287]  [ 10/312]  eta: 0:08:35  lr: 0.000022  min_lr: 0.000022  loss: 3.9554 (3.7764)  weight_decay: 0.0500 (0.0500)  time: 1.7078  data: 1.0304  max mem: 21002
Epoch: [287]  [ 20/312]  eta: 0:05:06  lr: 0.000022  min_lr: 0.000022  loss: 3.6063 (3.6179)  weight_decay: 0.0500 (0.0500)  time: 0.4885  data: 0.1279  max mem: 21002
Epoch: [287]  [ 30/312]  eta: 0:03:46  lr: 0.000022  min_lr: 0.000022  loss: 3.4851 (3.5027)  weight_decay: 0.0500 (0.0500)  time: 0.3068  data: 0.0116  max mem: 21002
Epoch: [287]  [ 40/312]  eta: 0:03:04  lr: 0.000022  min_lr: 0.000022  loss: 3.4921 (3.5213)  weight_decay: 0.0500 (0.0500)  time: 0.2899  data: 0.0008  max mem: 21002
Epoch: [287]  [ 50/312]  eta: 0:02:58  lr: 0.000022  min_lr: 0.000022  loss: 3.6435 (3.5403)  weight_decay: 0.0500 (0.0500)  time: 0.4915  data: 0.1580  max mem: 21002
Epoch: [287]  [ 60/312]  eta: 0:02:37  lr: 0.000022  min_lr: 0.000022  loss: 3.5056 (3.5300)  weight_decay: 0.0500 (0.0500)  time: 0.5084  data: 0.1730  max mem: 21002
Epoch: [287]  [ 70/312]  eta: 0:02:32  lr: 0.000022  min_lr: 0.000022  loss: 3.3590 (3.5175)  weight_decay: 0.0500 (0.0500)  time: 0.4944  data: 0.1622  max mem: 21002
Epoch: [287]  [ 80/312]  eta: 0:02:24  lr: 0.000021  min_lr: 0.000021  loss: 3.3590 (3.5246)  weight_decay: 0.0500 (0.0500)  time: 0.6185  data: 0.2244  max mem: 21002
Epoch: [287]  [ 90/312]  eta: 0:02:13  lr: 0.000021  min_lr: 0.000021  loss: 3.6739 (3.5421)  weight_decay: 0.0500 (0.0500)  time: 0.4947  data: 0.1390  max mem: 21002
Epoch: [287]  [100/312]  eta: 0:02:07  lr: 0.000021  min_lr: 0.000021  loss: 3.7961 (3.5671)  weight_decay: 0.0500 (0.0500)  time: 0.5224  data: 0.1899  max mem: 21002
Epoch: [287]  [110/312]  eta: 0:01:57  lr: 0.000021  min_lr: 0.000021  loss: 3.8261 (3.5784)  weight_decay: 0.0500 (0.0500)  time: 0.4913  data: 0.1456  max mem: 21002
Epoch: [287]  [120/312]  eta: 0:01:51  lr: 0.000021  min_lr: 0.000021  loss: 3.6979 (3.5858)  weight_decay: 0.0500 (0.0500)  time: 0.4848  data: 0.1693  max mem: 21002
Epoch: [287]  [130/312]  eta: 0:01:47  lr: 0.000021  min_lr: 0.000021  loss: 3.7343 (3.5918)  weight_decay: 0.0500 (0.0500)  time: 0.6595  data: 0.2836  max mem: 21002
Epoch: [287]  [140/312]  eta: 0:01:38  lr: 0.000021  min_lr: 0.000021  loss: 3.7561 (3.5788)  weight_decay: 0.0500 (0.0500)  time: 0.5226  data: 0.1365  max mem: 21002
Epoch: [287]  [150/312]  eta: 0:01:34  lr: 0.000021  min_lr: 0.000021  loss: 3.5806 (3.5714)  weight_decay: 0.0500 (0.0500)  time: 0.5052  data: 0.1361  max mem: 21002
Epoch: [287]  [160/312]  eta: 0:01:27  lr: 0.000021  min_lr: 0.000021  loss: 3.2235 (3.5432)  weight_decay: 0.0500 (0.0500)  time: 0.6036  data: 0.2399  max mem: 21002
Epoch: [287]  [170/312]  eta: 0:01:21  lr: 0.000020  min_lr: 0.000020  loss: 3.3367 (3.5498)  weight_decay: 0.0500 (0.0500)  time: 0.5232  data: 0.1321  max mem: 21002
Epoch: [287]  [180/312]  eta: 0:01:16  lr: 0.000020  min_lr: 0.000020  loss: 3.7342 (3.5541)  weight_decay: 0.0500 (0.0500)  time: 0.5643  data: 0.1231  max mem: 21002
Epoch: [287]  [190/312]  eta: 0:01:09  lr: 0.000020  min_lr: 0.000020  loss: 3.6911 (3.5419)  weight_decay: 0.0500 (0.0500)  time: 0.5032  data: 0.1052  max mem: 21002
Epoch: [287]  [200/312]  eta: 0:01:03  lr: 0.000020  min_lr: 0.000020  loss: 3.6911 (3.5465)  weight_decay: 0.0500 (0.0500)  time: 0.4997  data: 0.0632  max mem: 21002
Epoch: [287]  [210/312]  eta: 0:00:58  lr: 0.000020  min_lr: 0.000020  loss: 3.7322 (3.5477)  weight_decay: 0.0500 (0.0500)  time: 0.6323  data: 0.1116  max mem: 21002
Epoch: [287]  [220/312]  eta: 0:00:52  lr: 0.000020  min_lr: 0.000020  loss: 3.4025 (3.5373)  weight_decay: 0.0500 (0.0500)  time: 0.5569  data: 0.1046  max mem: 21002
Epoch: [287]  [230/312]  eta: 0:00:46  lr: 0.000020  min_lr: 0.000020  loss: 3.4109 (3.5377)  weight_decay: 0.0500 (0.0500)  time: 0.5347  data: 0.0694  max mem: 21002
Epoch: [287]  [240/312]  eta: 0:00:40  lr: 0.000020  min_lr: 0.000020  loss: 3.5785 (3.5432)  weight_decay: 0.0500 (0.0500)  time: 0.5111  data: 0.0659  max mem: 21002
Epoch: [287]  [250/312]  eta: 0:00:34  lr: 0.000020  min_lr: 0.000020  loss: 3.6747 (3.5472)  weight_decay: 0.0500 (0.0500)  time: 0.4881  data: 0.0986  max mem: 21002
Epoch: [287]  [260/312]  eta: 0:00:29  lr: 0.000020  min_lr: 0.000020  loss: 3.6747 (3.5505)  weight_decay: 0.0500 (0.0500)  time: 0.5863  data: 0.0854  max mem: 21002
Epoch: [287]  [270/312]  eta: 0:00:23  lr: 0.000020  min_lr: 0.000020  loss: 3.6700 (3.5483)  weight_decay: 0.0500 (0.0500)  time: 0.5629  data: 0.0746  max mem: 21002
Epoch: [287]  [280/312]  eta: 0:00:17  lr: 0.000019  min_lr: 0.000019  loss: 3.6700 (3.5518)  weight_decay: 0.0500 (0.0500)  time: 0.4917  data: 0.0998  max mem: 21002
Epoch: [287]  [290/312]  eta: 0:00:12  lr: 0.000019  min_lr: 0.000019  loss: 3.6642 (3.5543)  weight_decay: 0.0500 (0.0500)  time: 0.5700  data: 0.1552  max mem: 21002
Epoch: [287]  [300/312]  eta: 0:00:06  lr: 0.000019  min_lr: 0.000019  loss: 3.8229 (3.5681)  weight_decay: 0.0500 (0.0500)  time: 0.4738  data: 0.0960  max mem: 21002
Epoch: [287]  [310/312]  eta: 0:00:01  lr: 0.000019  min_lr: 0.000019  loss: 3.7843 (3.5658)  weight_decay: 0.0500 (0.0500)  time: 0.2797  data: 0.0001  max mem: 21002
Epoch: [287]  [311/312]  eta: 0:00:00  lr: 0.000019  min_lr: 0.000019  loss: 3.6389 (3.5646)  weight_decay: 0.0500 (0.0500)  time: 0.2796  data: 0.0001  max mem: 21002
Epoch: [287] Total time: 0:02:50 (0.5481 s / it)
Averaged stats: lr: 0.000019  min_lr: 0.000019  loss: 3.6389 (3.5620)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.0420 (1.0420)  acc1: 81.2500 (81.2500)  acc5: 95.5729 (95.5729)  time: 8.3554  data: 8.2365  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3628 (1.2812)  acc1: 73.1771 (74.1120)  acc5: 91.7969 (92.0480)  time: 1.0157  data: 0.9153  max mem: 21002
Test: Total time: 0:00:09 (1.0261 s / it)
* Acc@1 73.902 Acc@5 92.110 loss 1.286
Accuracy of the model on the 50000 test images: 73.9%
Max accuracy: 73.99%
Epoch: [288]  [  0/312]  eta: 0:59:03  lr: 0.000019  min_lr: 0.000019  loss: 4.3655 (4.3655)  weight_decay: 0.0500 (0.0500)  time: 11.3578  data: 9.2528  max mem: 21002
Epoch: [288]  [ 10/312]  eta: 0:08:41  lr: 0.000019  min_lr: 0.000019  loss: 3.8909 (3.7931)  weight_decay: 0.0500 (0.0500)  time: 1.7283  data: 1.0727  max mem: 21002
Epoch: [288]  [ 20/312]  eta: 0:05:06  lr: 0.000019  min_lr: 0.000019  loss: 3.8088 (3.7324)  weight_decay: 0.0500 (0.0500)  time: 0.5325  data: 0.1320  max mem: 21002
Epoch: [288]  [ 30/312]  eta: 0:03:46  lr: 0.000019  min_lr: 0.000019  loss: 3.7128 (3.7230)  weight_decay: 0.0500 (0.0500)  time: 0.2950  data: 0.0050  max mem: 21002
Epoch: [288]  [ 40/312]  eta: 0:03:04  lr: 0.000019  min_lr: 0.000019  loss: 3.7508 (3.7204)  weight_decay: 0.0500 (0.0500)  time: 0.2884  data: 0.0008  max mem: 21002
Epoch: [288]  [ 50/312]  eta: 0:02:53  lr: 0.000019  min_lr: 0.000019  loss: 3.8521 (3.7168)  weight_decay: 0.0500 (0.0500)  time: 0.4449  data: 0.1090  max mem: 21002
Epoch: [288]  [ 60/312]  eta: 0:02:34  lr: 0.000019  min_lr: 0.000019  loss: 3.7568 (3.7024)  weight_decay: 0.0500 (0.0500)  time: 0.4849  data: 0.1472  max mem: 21002
Epoch: [288]  [ 70/312]  eta: 0:02:28  lr: 0.000018  min_lr: 0.000018  loss: 3.6901 (3.6790)  weight_decay: 0.0500 (0.0500)  time: 0.4883  data: 0.1270  max mem: 21002
Epoch: [288]  [ 80/312]  eta: 0:02:20  lr: 0.000018  min_lr: 0.000018  loss: 3.6357 (3.6802)  weight_decay: 0.0500 (0.0500)  time: 0.5874  data: 0.1239  max mem: 21002
Epoch: [288]  [ 90/312]  eta: 0:02:11  lr: 0.000018  min_lr: 0.000018  loss: 3.7091 (3.6788)  weight_decay: 0.0500 (0.0500)  time: 0.5163  data: 0.0899  max mem: 21002
Epoch: [288]  [100/312]  eta: 0:02:06  lr: 0.000018  min_lr: 0.000018  loss: 3.5397 (3.6419)  weight_decay: 0.0500 (0.0500)  time: 0.5539  data: 0.1343  max mem: 21002
Epoch: [288]  [110/312]  eta: 0:01:56  lr: 0.000018  min_lr: 0.000018  loss: 3.6044 (3.6462)  weight_decay: 0.0500 (0.0500)  time: 0.4940  data: 0.0915  max mem: 21002
Epoch: [288]  [120/312]  eta: 0:01:51  lr: 0.000018  min_lr: 0.000018  loss: 3.7210 (3.6373)  weight_decay: 0.0500 (0.0500)  time: 0.4963  data: 0.0782  max mem: 21002
Epoch: [288]  [130/312]  eta: 0:01:45  lr: 0.000018  min_lr: 0.000018  loss: 3.7210 (3.6355)  weight_decay: 0.0500 (0.0500)  time: 0.6249  data: 0.1241  max mem: 21002
Epoch: [288]  [140/312]  eta: 0:01:37  lr: 0.000018  min_lr: 0.000018  loss: 3.7812 (3.6366)  weight_decay: 0.0500 (0.0500)  time: 0.5019  data: 0.0842  max mem: 21002
Epoch: [288]  [150/312]  eta: 0:01:33  lr: 0.000018  min_lr: 0.000018  loss: 3.6118 (3.6241)  weight_decay: 0.0500 (0.0500)  time: 0.5306  data: 0.0963  max mem: 21002
Epoch: [288]  [160/312]  eta: 0:01:26  lr: 0.000018  min_lr: 0.000018  loss: 3.5830 (3.6175)  weight_decay: 0.0500 (0.0500)  time: 0.5829  data: 0.1046  max mem: 21002
Epoch: [288]  [170/312]  eta: 0:01:20  lr: 0.000017  min_lr: 0.000017  loss: 3.7800 (3.6266)  weight_decay: 0.0500 (0.0500)  time: 0.4759  data: 0.0981  max mem: 21002
Epoch: [288]  [180/312]  eta: 0:01:14  lr: 0.000017  min_lr: 0.000017  loss: 3.8746 (3.6345)  weight_decay: 0.0500 (0.0500)  time: 0.5418  data: 0.1408  max mem: 21002
Epoch: [288]  [190/312]  eta: 0:01:08  lr: 0.000017  min_lr: 0.000017  loss: 3.8999 (3.6330)  weight_decay: 0.0500 (0.0500)  time: 0.5213  data: 0.1146  max mem: 21002
Epoch: [288]  [200/312]  eta: 0:01:03  lr: 0.000017  min_lr: 0.000017  loss: 3.8232 (3.6414)  weight_decay: 0.0500 (0.0500)  time: 0.5265  data: 0.1062  max mem: 21002
Epoch: [288]  [210/312]  eta: 0:00:57  lr: 0.000017  min_lr: 0.000017  loss: 3.8950 (3.6543)  weight_decay: 0.0500 (0.0500)  time: 0.6494  data: 0.1338  max mem: 21002
Epoch: [288]  [220/312]  eta: 0:00:51  lr: 0.000017  min_lr: 0.000017  loss: 3.6414 (3.6431)  weight_decay: 0.0500 (0.0500)  time: 0.5226  data: 0.1172  max mem: 21002
Epoch: [288]  [230/312]  eta: 0:00:46  lr: 0.000017  min_lr: 0.000017  loss: 3.4588 (3.6323)  weight_decay: 0.0500 (0.0500)  time: 0.5014  data: 0.1169  max mem: 21002
Epoch: [288]  [240/312]  eta: 0:00:40  lr: 0.000017  min_lr: 0.000017  loss: 3.4375 (3.6289)  weight_decay: 0.0500 (0.0500)  time: 0.5273  data: 0.0880  max mem: 21002
Epoch: [288]  [250/312]  eta: 0:00:34  lr: 0.000017  min_lr: 0.000017  loss: 3.6096 (3.6314)  weight_decay: 0.0500 (0.0500)  time: 0.5211  data: 0.1183  max mem: 21002
Epoch: [288]  [260/312]  eta: 0:00:29  lr: 0.000017  min_lr: 0.000017  loss: 3.7015 (3.6288)  weight_decay: 0.0500 (0.0500)  time: 0.6154  data: 0.1572  max mem: 21002
Epoch: [288]  [270/312]  eta: 0:00:23  lr: 0.000017  min_lr: 0.000017  loss: 3.8391 (3.6309)  weight_decay: 0.0500 (0.0500)  time: 0.5048  data: 0.1105  max mem: 21002
Epoch: [288]  [280/312]  eta: 0:00:17  lr: 0.000016  min_lr: 0.000016  loss: 3.8168 (3.6245)  weight_decay: 0.0500 (0.0500)  time: 0.5047  data: 0.1160  max mem: 21002
Epoch: [288]  [290/312]  eta: 0:00:12  lr: 0.000016  min_lr: 0.000016  loss: 3.5813 (3.6257)  weight_decay: 0.0500 (0.0500)  time: 0.6265  data: 0.1354  max mem: 21002
Epoch: [288]  [300/312]  eta: 0:00:06  lr: 0.000016  min_lr: 0.000016  loss: 3.6487 (3.6213)  weight_decay: 0.0500 (0.0500)  time: 0.4776  data: 0.0966  max mem: 21002
Epoch: [288]  [310/312]  eta: 0:00:01  lr: 0.000016  min_lr: 0.000016  loss: 3.8383 (3.6293)  weight_decay: 0.0500 (0.0500)  time: 0.3058  data: 0.0295  max mem: 21002
Epoch: [288]  [311/312]  eta: 0:00:00  lr: 0.000016  min_lr: 0.000016  loss: 3.8383 (3.6263)  weight_decay: 0.0500 (0.0500)  time: 0.2838  data: 0.0070  max mem: 21002
Epoch: [288] Total time: 0:02:50 (0.5462 s / it)
Averaged stats: lr: 0.000016  min_lr: 0.000016  loss: 3.8383 (3.5739)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:21  loss: 1.0607 (1.0607)  acc1: 80.7292 (80.7292)  acc5: 95.5729 (95.5729)  time: 9.0358  data: 8.9172  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3802 (1.2960)  acc1: 73.0469 (73.6640)  acc5: 90.7552 (91.8080)  time: 1.0907  data: 0.9909  max mem: 21002
Test: Total time: 0:00:10 (1.1204 s / it)
* Acc@1 73.712 Acc@5 91.964 loss 1.300
Accuracy of the model on the 50000 test images: 73.7%
Max accuracy: 73.99%
Epoch: [289]  [  0/312]  eta: 1:03:17  lr: 0.000016  min_lr: 0.000016  loss: 2.9109 (2.9109)  weight_decay: 0.0500 (0.0500)  time: 12.1705  data: 11.2876  max mem: 21002
Epoch: [289]  [ 10/312]  eta: 0:07:27  lr: 0.000016  min_lr: 0.000016  loss: 3.4067 (3.4479)  weight_decay: 0.0500 (0.0500)  time: 1.4821  data: 1.0267  max mem: 21002
Epoch: [289]  [ 20/312]  eta: 0:05:12  lr: 0.000016  min_lr: 0.000016  loss: 3.4067 (3.5093)  weight_decay: 0.0500 (0.0500)  time: 0.5159  data: 0.0999  max mem: 21002
Epoch: [289]  [ 30/312]  eta: 0:03:51  lr: 0.000016  min_lr: 0.000016  loss: 3.8282 (3.5383)  weight_decay: 0.0500 (0.0500)  time: 0.4551  data: 0.1009  max mem: 21002
Epoch: [289]  [ 40/312]  eta: 0:03:36  lr: 0.000016  min_lr: 0.000016  loss: 3.6931 (3.5066)  weight_decay: 0.0500 (0.0500)  time: 0.5091  data: 0.0733  max mem: 21002
Epoch: [289]  [ 50/312]  eta: 0:03:20  lr: 0.000016  min_lr: 0.000016  loss: 3.4041 (3.4937)  weight_decay: 0.0500 (0.0500)  time: 0.6781  data: 0.1075  max mem: 21002
Epoch: [289]  [ 60/312]  eta: 0:02:53  lr: 0.000016  min_lr: 0.000016  loss: 3.6393 (3.5159)  weight_decay: 0.0500 (0.0500)  time: 0.4698  data: 0.0472  max mem: 21002
Epoch: [289]  [ 70/312]  eta: 0:02:45  lr: 0.000016  min_lr: 0.000016  loss: 3.5013 (3.5129)  weight_decay: 0.0500 (0.0500)  time: 0.4778  data: 0.0512  max mem: 21002
Epoch: [289]  [ 80/312]  eta: 0:02:35  lr: 0.000016  min_lr: 0.000016  loss: 3.4454 (3.5089)  weight_decay: 0.0500 (0.0500)  time: 0.6061  data: 0.0752  max mem: 21002
Epoch: [289]  [ 90/312]  eta: 0:02:21  lr: 0.000015  min_lr: 0.000015  loss: 3.5099 (3.5086)  weight_decay: 0.0500 (0.0500)  time: 0.4769  data: 0.0671  max mem: 21002
Epoch: [289]  [100/312]  eta: 0:02:15  lr: 0.000015  min_lr: 0.000015  loss: 3.5099 (3.4984)  weight_decay: 0.0500 (0.0500)  time: 0.5175  data: 0.0982  max mem: 21002
Epoch: [289]  [110/312]  eta: 0:02:02  lr: 0.000015  min_lr: 0.000015  loss: 3.3931 (3.4832)  weight_decay: 0.0500 (0.0500)  time: 0.4698  data: 0.0673  max mem: 21002
Epoch: [289]  [120/312]  eta: 0:01:57  lr: 0.000015  min_lr: 0.000015  loss: 3.7355 (3.5126)  weight_decay: 0.0500 (0.0500)  time: 0.4724  data: 0.0686  max mem: 21002
Epoch: [289]  [130/312]  eta: 0:01:51  lr: 0.000015  min_lr: 0.000015  loss: 3.8776 (3.5328)  weight_decay: 0.0500 (0.0500)  time: 0.6569  data: 0.1492  max mem: 21002
Epoch: [289]  [140/312]  eta: 0:01:41  lr: 0.000015  min_lr: 0.000015  loss: 3.7904 (3.5399)  weight_decay: 0.0500 (0.0500)  time: 0.4742  data: 0.0814  max mem: 21002
Epoch: [289]  [150/312]  eta: 0:01:36  lr: 0.000015  min_lr: 0.000015  loss: 3.5807 (3.5358)  weight_decay: 0.0500 (0.0500)  time: 0.4753  data: 0.1094  max mem: 21002
Epoch: [289]  [160/312]  eta: 0:01:31  lr: 0.000015  min_lr: 0.000015  loss: 3.4096 (3.5307)  weight_decay: 0.0500 (0.0500)  time: 0.6515  data: 0.2503  max mem: 21002
Epoch: [289]  [170/312]  eta: 0:01:22  lr: 0.000015  min_lr: 0.000015  loss: 3.3431 (3.5247)  weight_decay: 0.0500 (0.0500)  time: 0.4742  data: 0.1416  max mem: 21002
Epoch: [289]  [180/312]  eta: 0:01:17  lr: 0.000015  min_lr: 0.000015  loss: 3.3135 (3.5251)  weight_decay: 0.0500 (0.0500)  time: 0.4770  data: 0.1499  max mem: 21002
Epoch: [289]  [190/312]  eta: 0:01:09  lr: 0.000015  min_lr: 0.000015  loss: 3.8012 (3.5465)  weight_decay: 0.0500 (0.0500)  time: 0.4670  data: 0.1499  max mem: 21002
Epoch: [289]  [200/312]  eta: 0:01:04  lr: 0.000014  min_lr: 0.000014  loss: 3.8899 (3.5518)  weight_decay: 0.0500 (0.0500)  time: 0.4848  data: 0.1656  max mem: 21002
Epoch: [289]  [210/312]  eta: 0:00:59  lr: 0.000014  min_lr: 0.000014  loss: 3.5858 (3.5451)  weight_decay: 0.0500 (0.0500)  time: 0.6818  data: 0.3264  max mem: 21002
Epoch: [289]  [220/312]  eta: 0:00:52  lr: 0.000014  min_lr: 0.000014  loss: 3.6730 (3.5465)  weight_decay: 0.0500 (0.0500)  time: 0.4845  data: 0.1614  max mem: 21002
Epoch: [289]  [230/312]  eta: 0:00:47  lr: 0.000014  min_lr: 0.000014  loss: 3.7012 (3.5487)  weight_decay: 0.0500 (0.0500)  time: 0.5014  data: 0.1355  max mem: 21002
Epoch: [289]  [240/312]  eta: 0:00:41  lr: 0.000014  min_lr: 0.000014  loss: 3.6733 (3.5465)  weight_decay: 0.0500 (0.0500)  time: 0.6688  data: 0.2584  max mem: 21002
Epoch: [289]  [250/312]  eta: 0:00:35  lr: 0.000014  min_lr: 0.000014  loss: 3.4222 (3.5390)  weight_decay: 0.0500 (0.0500)  time: 0.4761  data: 0.1236  max mem: 21002
Epoch: [289]  [260/312]  eta: 0:00:29  lr: 0.000014  min_lr: 0.000014  loss: 3.3727 (3.5348)  weight_decay: 0.0500 (0.0500)  time: 0.4851  data: 0.1434  max mem: 21002
Epoch: [289]  [270/312]  eta: 0:00:23  lr: 0.000014  min_lr: 0.000014  loss: 3.6829 (3.5415)  weight_decay: 0.0500 (0.0500)  time: 0.4653  data: 0.1433  max mem: 21002
Epoch: [289]  [280/312]  eta: 0:00:18  lr: 0.000014  min_lr: 0.000014  loss: 3.8095 (3.5499)  weight_decay: 0.0500 (0.0500)  time: 0.5344  data: 0.1049  max mem: 21002
Epoch: [289]  [290/312]  eta: 0:00:12  lr: 0.000014  min_lr: 0.000014  loss: 3.7437 (3.5530)  weight_decay: 0.0500 (0.0500)  time: 0.6427  data: 0.1492  max mem: 21002
Epoch: [289]  [300/312]  eta: 0:00:06  lr: 0.000014  min_lr: 0.000014  loss: 3.5571 (3.5506)  weight_decay: 0.0500 (0.0500)  time: 0.3926  data: 0.0447  max mem: 21002
Epoch: [289]  [310/312]  eta: 0:00:01  lr: 0.000014  min_lr: 0.000014  loss: 3.4506 (3.5499)  weight_decay: 0.0500 (0.0500)  time: 0.2772  data: 0.0001  max mem: 21002
Epoch: [289]  [311/312]  eta: 0:00:00  lr: 0.000014  min_lr: 0.000014  loss: 3.4506 (3.5522)  weight_decay: 0.0500 (0.0500)  time: 0.2771  data: 0.0001  max mem: 21002
Epoch: [289] Total time: 0:02:51 (0.5491 s / it)
Averaged stats: lr: 0.000014  min_lr: 0.000014  loss: 3.4506 (3.5817)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:11  loss: 1.0493 (1.0493)  acc1: 81.3802 (81.3802)  acc5: 96.0938 (96.0938)  time: 7.9220  data: 7.8037  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3745 (1.2923)  acc1: 73.5677 (74.0960)  acc5: 91.7969 (92.2720)  time: 1.0288  data: 0.9318  max mem: 21002
Test: Total time: 0:00:09 (1.0412 s / it)
* Acc@1 73.980 Acc@5 92.208 loss 1.296
Accuracy of the model on the 50000 test images: 74.0%
Max accuracy: 73.99%
Epoch: [290]  [  0/312]  eta: 1:04:33  lr: 0.000014  min_lr: 0.000014  loss: 2.4894 (2.4894)  weight_decay: 0.0500 (0.0500)  time: 12.4137  data: 11.9590  max mem: 21002
Epoch: [290]  [ 10/312]  eta: 0:08:09  lr: 0.000013  min_lr: 0.000013  loss: 3.7838 (3.5209)  weight_decay: 0.0500 (0.0500)  time: 1.6213  data: 1.0881  max mem: 21002
Epoch: [290]  [ 20/312]  eta: 0:05:09  lr: 0.000013  min_lr: 0.000013  loss: 3.7838 (3.5952)  weight_decay: 0.0500 (0.0500)  time: 0.4918  data: 0.0185  max mem: 21002
Epoch: [290]  [ 30/312]  eta: 0:03:49  lr: 0.000013  min_lr: 0.000013  loss: 3.5239 (3.6099)  weight_decay: 0.0500 (0.0500)  time: 0.3674  data: 0.0183  max mem: 21002
Epoch: [290]  [ 40/312]  eta: 0:03:17  lr: 0.000013  min_lr: 0.000013  loss: 3.6956 (3.6243)  weight_decay: 0.0500 (0.0500)  time: 0.3743  data: 0.0569  max mem: 21002
Epoch: [290]  [ 50/312]  eta: 0:03:05  lr: 0.000013  min_lr: 0.000013  loss: 3.6932 (3.5951)  weight_decay: 0.0500 (0.0500)  time: 0.5462  data: 0.1549  max mem: 21002
Epoch: [290]  [ 60/312]  eta: 0:02:41  lr: 0.000013  min_lr: 0.000013  loss: 3.6874 (3.6230)  weight_decay: 0.0500 (0.0500)  time: 0.4621  data: 0.0987  max mem: 21002
Epoch: [290]  [ 70/312]  eta: 0:02:35  lr: 0.000013  min_lr: 0.000013  loss: 3.9144 (3.6678)  weight_decay: 0.0500 (0.0500)  time: 0.4785  data: 0.1328  max mem: 21002
Epoch: [290]  [ 80/312]  eta: 0:02:27  lr: 0.000013  min_lr: 0.000013  loss: 3.7376 (3.6387)  weight_decay: 0.0500 (0.0500)  time: 0.6340  data: 0.2504  max mem: 21002
Epoch: [290]  [ 90/312]  eta: 0:02:14  lr: 0.000013  min_lr: 0.000013  loss: 3.8655 (3.6732)  weight_decay: 0.0500 (0.0500)  time: 0.4716  data: 0.1378  max mem: 21002
Epoch: [290]  [100/312]  eta: 0:02:09  lr: 0.000013  min_lr: 0.000013  loss: 3.8655 (3.6809)  weight_decay: 0.0500 (0.0500)  time: 0.5011  data: 0.1553  max mem: 21002
Epoch: [290]  [110/312]  eta: 0:01:57  lr: 0.000013  min_lr: 0.000013  loss: 3.7939 (3.6684)  weight_decay: 0.0500 (0.0500)  time: 0.4726  data: 0.1361  max mem: 21002
Epoch: [290]  [120/312]  eta: 0:01:53  lr: 0.000013  min_lr: 0.000013  loss: 3.5605 (3.6439)  weight_decay: 0.0500 (0.0500)  time: 0.4880  data: 0.1581  max mem: 21002
Epoch: [290]  [130/312]  eta: 0:01:48  lr: 0.000013  min_lr: 0.000013  loss: 3.5605 (3.6302)  weight_decay: 0.0500 (0.0500)  time: 0.6744  data: 0.2973  max mem: 21002
Epoch: [290]  [140/312]  eta: 0:01:38  lr: 0.000012  min_lr: 0.000012  loss: 3.6545 (3.6209)  weight_decay: 0.0500 (0.0500)  time: 0.4743  data: 0.1403  max mem: 21002
Epoch: [290]  [150/312]  eta: 0:01:33  lr: 0.000012  min_lr: 0.000012  loss: 3.6054 (3.6199)  weight_decay: 0.0500 (0.0500)  time: 0.4701  data: 0.1388  max mem: 21002
Epoch: [290]  [160/312]  eta: 0:01:28  lr: 0.000012  min_lr: 0.000012  loss: 3.5923 (3.6138)  weight_decay: 0.0500 (0.0500)  time: 0.6413  data: 0.2623  max mem: 21002
Epoch: [290]  [170/312]  eta: 0:01:20  lr: 0.000012  min_lr: 0.000012  loss: 3.6200 (3.6035)  weight_decay: 0.0500 (0.0500)  time: 0.4926  data: 0.1588  max mem: 21002
Epoch: [290]  [180/312]  eta: 0:01:16  lr: 0.000012  min_lr: 0.000012  loss: 3.8070 (3.6223)  weight_decay: 0.0500 (0.0500)  time: 0.5651  data: 0.1766  max mem: 21002
Epoch: [290]  [190/312]  eta: 0:01:08  lr: 0.000012  min_lr: 0.000012  loss: 3.8695 (3.6143)  weight_decay: 0.0500 (0.0500)  time: 0.5314  data: 0.1423  max mem: 21002
Epoch: [290]  [200/312]  eta: 0:01:03  lr: 0.000012  min_lr: 0.000012  loss: 3.3702 (3.6007)  weight_decay: 0.0500 (0.0500)  time: 0.4704  data: 0.1053  max mem: 21002
Epoch: [290]  [210/312]  eta: 0:00:58  lr: 0.000012  min_lr: 0.000012  loss: 3.4775 (3.5974)  weight_decay: 0.0500 (0.0500)  time: 0.6611  data: 0.1988  max mem: 21002
Epoch: [290]  [220/312]  eta: 0:00:51  lr: 0.000012  min_lr: 0.000012  loss: 3.4775 (3.5863)  weight_decay: 0.0500 (0.0500)  time: 0.4774  data: 0.0945  max mem: 21002
Epoch: [290]  [230/312]  eta: 0:00:46  lr: 0.000012  min_lr: 0.000012  loss: 3.4717 (3.5827)  weight_decay: 0.0500 (0.0500)  time: 0.5014  data: 0.1049  max mem: 21002
Epoch: [290]  [240/312]  eta: 0:00:40  lr: 0.000012  min_lr: 0.000012  loss: 3.6415 (3.5776)  weight_decay: 0.0500 (0.0500)  time: 0.6560  data: 0.1682  max mem: 21002
Epoch: [290]  [250/312]  eta: 0:00:34  lr: 0.000012  min_lr: 0.000012  loss: 3.6626 (3.5866)  weight_decay: 0.0500 (0.0500)  time: 0.4969  data: 0.1085  max mem: 21002
Epoch: [290]  [260/312]  eta: 0:00:29  lr: 0.000012  min_lr: 0.000012  loss: 3.7569 (3.5834)  weight_decay: 0.0500 (0.0500)  time: 0.5605  data: 0.1448  max mem: 21002
Epoch: [290]  [270/312]  eta: 0:00:23  lr: 0.000011  min_lr: 0.000011  loss: 3.8322 (3.5919)  weight_decay: 0.0500 (0.0500)  time: 0.5052  data: 0.1005  max mem: 21002
Epoch: [290]  [280/312]  eta: 0:00:18  lr: 0.000011  min_lr: 0.000011  loss: 3.7911 (3.5923)  weight_decay: 0.0500 (0.0500)  time: 0.5259  data: 0.0747  max mem: 21002
Epoch: [290]  [290/312]  eta: 0:00:12  lr: 0.000011  min_lr: 0.000011  loss: 3.8712 (3.6065)  weight_decay: 0.0500 (0.0500)  time: 0.6368  data: 0.0860  max mem: 21002
Epoch: [290]  [300/312]  eta: 0:00:06  lr: 0.000011  min_lr: 0.000011  loss: 3.8712 (3.6131)  weight_decay: 0.0500 (0.0500)  time: 0.3940  data: 0.0116  max mem: 21002
Epoch: [290]  [310/312]  eta: 0:00:01  lr: 0.000011  min_lr: 0.000011  loss: 3.5296 (3.5998)  weight_decay: 0.0500 (0.0500)  time: 0.2834  data: 0.0001  max mem: 21002
Epoch: [290]  [311/312]  eta: 0:00:00  lr: 0.000011  min_lr: 0.000011  loss: 3.5429 (3.6008)  weight_decay: 0.0500 (0.0500)  time: 0.2832  data: 0.0001  max mem: 21002
Epoch: [290] Total time: 0:02:51 (0.5482 s / it)
Averaged stats: lr: 0.000011  min_lr: 0.000011  loss: 3.5429 (3.5650)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 1.0031 (1.0031)  acc1: 81.6406 (81.6406)  acc5: 96.0938 (96.0938)  time: 8.1188  data: 8.0001  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3290 (1.2501)  acc1: 73.1771 (74.1760)  acc5: 92.1875 (92.2880)  time: 1.0023  data: 0.9053  max mem: 21002
Test: Total time: 0:00:09 (1.0135 s / it)
* Acc@1 74.030 Acc@5 92.196 loss 1.255
Accuracy of the model on the 50000 test images: 74.0%
Max accuracy: 74.03%
Epoch: [291]  [  0/312]  eta: 1:04:32  lr: 0.000011  min_lr: 0.000011  loss: 3.0437 (3.0437)  weight_decay: 0.0500 (0.0500)  time: 12.4116  data: 12.1098  max mem: 21002
Epoch: [291]  [ 10/312]  eta: 0:08:08  lr: 0.000011  min_lr: 0.000011  loss: 3.7049 (3.6954)  weight_decay: 0.0500 (0.0500)  time: 1.6165  data: 1.1016  max mem: 21002
Epoch: [291]  [ 20/312]  eta: 0:05:03  lr: 0.000011  min_lr: 0.000011  loss: 3.7936 (3.7102)  weight_decay: 0.0500 (0.0500)  time: 0.4698  data: 0.0607  max mem: 21002
Epoch: [291]  [ 30/312]  eta: 0:03:44  lr: 0.000011  min_lr: 0.000011  loss: 3.7260 (3.6481)  weight_decay: 0.0500 (0.0500)  time: 0.3473  data: 0.0608  max mem: 21002
Epoch: [291]  [ 40/312]  eta: 0:03:14  lr: 0.000011  min_lr: 0.000011  loss: 3.7703 (3.7093)  weight_decay: 0.0500 (0.0500)  time: 0.3757  data: 0.0198  max mem: 21002
Epoch: [291]  [ 50/312]  eta: 0:03:05  lr: 0.000011  min_lr: 0.000011  loss: 3.8269 (3.6862)  weight_decay: 0.0500 (0.0500)  time: 0.5645  data: 0.1129  max mem: 21002
Epoch: [291]  [ 60/312]  eta: 0:02:40  lr: 0.000011  min_lr: 0.000011  loss: 3.6927 (3.6605)  weight_decay: 0.0500 (0.0500)  time: 0.4816  data: 0.0986  max mem: 21002
Epoch: [291]  [ 70/312]  eta: 0:02:37  lr: 0.000011  min_lr: 0.000011  loss: 3.3646 (3.5777)  weight_decay: 0.0500 (0.0500)  time: 0.5158  data: 0.1047  max mem: 21002
Epoch: [291]  [ 80/312]  eta: 0:02:27  lr: 0.000011  min_lr: 0.000011  loss: 3.2948 (3.5515)  weight_decay: 0.0500 (0.0500)  time: 0.6349  data: 0.1445  max mem: 21002
Epoch: [291]  [ 90/312]  eta: 0:02:17  lr: 0.000011  min_lr: 0.000011  loss: 3.6828 (3.5700)  weight_decay: 0.0500 (0.0500)  time: 0.4961  data: 0.0793  max mem: 21002
Epoch: [291]  [100/312]  eta: 0:02:12  lr: 0.000010  min_lr: 0.000010  loss: 3.6618 (3.5718)  weight_decay: 0.0500 (0.0500)  time: 0.5623  data: 0.0752  max mem: 21002
Epoch: [291]  [110/312]  eta: 0:02:00  lr: 0.000010  min_lr: 0.000010  loss: 3.5835 (3.5681)  weight_decay: 0.0500 (0.0500)  time: 0.4905  data: 0.0565  max mem: 21002
Epoch: [291]  [120/312]  eta: 0:01:54  lr: 0.000010  min_lr: 0.000010  loss: 3.8134 (3.5716)  weight_decay: 0.0500 (0.0500)  time: 0.4648  data: 0.0696  max mem: 21002
Epoch: [291]  [130/312]  eta: 0:01:49  lr: 0.000010  min_lr: 0.000010  loss: 3.6050 (3.5674)  weight_decay: 0.0500 (0.0500)  time: 0.6270  data: 0.1042  max mem: 21002
Epoch: [291]  [140/312]  eta: 0:01:40  lr: 0.000010  min_lr: 0.000010  loss: 3.5592 (3.5538)  weight_decay: 0.0500 (0.0500)  time: 0.5014  data: 0.0887  max mem: 21002
Epoch: [291]  [150/312]  eta: 0:01:35  lr: 0.000010  min_lr: 0.000010  loss: 3.3382 (3.5426)  weight_decay: 0.0500 (0.0500)  time: 0.5193  data: 0.1121  max mem: 21002
Epoch: [291]  [160/312]  eta: 0:01:28  lr: 0.000010  min_lr: 0.000010  loss: 3.4698 (3.5489)  weight_decay: 0.0500 (0.0500)  time: 0.5825  data: 0.1060  max mem: 21002
Epoch: [291]  [170/312]  eta: 0:01:21  lr: 0.000010  min_lr: 0.000010  loss: 3.8134 (3.5703)  weight_decay: 0.0500 (0.0500)  time: 0.4804  data: 0.0632  max mem: 21002
Epoch: [291]  [180/312]  eta: 0:01:16  lr: 0.000010  min_lr: 0.000010  loss: 3.8134 (3.5675)  weight_decay: 0.0500 (0.0500)  time: 0.5434  data: 0.0755  max mem: 21002
Epoch: [291]  [190/312]  eta: 0:01:09  lr: 0.000010  min_lr: 0.000010  loss: 3.5434 (3.5695)  weight_decay: 0.0500 (0.0500)  time: 0.4796  data: 0.0742  max mem: 21002
Epoch: [291]  [200/312]  eta: 0:01:04  lr: 0.000010  min_lr: 0.000010  loss: 3.5434 (3.5734)  weight_decay: 0.0500 (0.0500)  time: 0.5599  data: 0.1102  max mem: 21002
Epoch: [291]  [210/312]  eta: 0:00:58  lr: 0.000010  min_lr: 0.000010  loss: 3.5431 (3.5743)  weight_decay: 0.0500 (0.0500)  time: 0.6407  data: 0.0810  max mem: 21002
Epoch: [291]  [220/312]  eta: 0:00:52  lr: 0.000010  min_lr: 0.000010  loss: 3.6033 (3.5795)  weight_decay: 0.0500 (0.0500)  time: 0.4612  data: 0.0172  max mem: 21002
Epoch: [291]  [230/312]  eta: 0:00:47  lr: 0.000010  min_lr: 0.000010  loss: 3.7675 (3.5847)  weight_decay: 0.0500 (0.0500)  time: 0.6054  data: 0.0596  max mem: 21002
Epoch: [291]  [240/312]  eta: 0:00:40  lr: 0.000010  min_lr: 0.000010  loss: 3.6715 (3.5819)  weight_decay: 0.0500 (0.0500)  time: 0.5934  data: 0.0439  max mem: 21002
Epoch: [291]  [250/312]  eta: 0:00:35  lr: 0.000009  min_lr: 0.000009  loss: 3.7560 (3.5925)  weight_decay: 0.0500 (0.0500)  time: 0.4654  data: 0.0288  max mem: 21002
Epoch: [291]  [260/312]  eta: 0:00:29  lr: 0.000009  min_lr: 0.000009  loss: 3.8558 (3.5991)  weight_decay: 0.0500 (0.0500)  time: 0.5993  data: 0.0285  max mem: 21002
Epoch: [291]  [270/312]  eta: 0:00:23  lr: 0.000009  min_lr: 0.000009  loss: 3.7054 (3.5980)  weight_decay: 0.0500 (0.0500)  time: 0.4915  data: 0.0184  max mem: 21002
Epoch: [291]  [280/312]  eta: 0:00:18  lr: 0.000009  min_lr: 0.000009  loss: 3.8465 (3.6044)  weight_decay: 0.0500 (0.0500)  time: 0.5248  data: 0.0525  max mem: 21002
Epoch: [291]  [290/312]  eta: 0:00:12  lr: 0.000009  min_lr: 0.000009  loss: 3.7465 (3.5993)  weight_decay: 0.0500 (0.0500)  time: 0.5927  data: 0.0351  max mem: 21002
Epoch: [291]  [300/312]  eta: 0:00:06  lr: 0.000009  min_lr: 0.000009  loss: 3.2324 (3.5865)  weight_decay: 0.0500 (0.0500)  time: 0.3889  data: 0.0136  max mem: 21002
Epoch: [291]  [310/312]  eta: 0:00:01  lr: 0.000009  min_lr: 0.000009  loss: 3.5151 (3.5916)  weight_decay: 0.0500 (0.0500)  time: 0.2975  data: 0.0135  max mem: 21002
Epoch: [291]  [311/312]  eta: 0:00:00  lr: 0.000009  min_lr: 0.000009  loss: 3.5595 (3.5915)  weight_decay: 0.0500 (0.0500)  time: 0.2901  data: 0.0135  max mem: 21002
Epoch: [291] Total time: 0:02:51 (0.5498 s / it)
Averaged stats: lr: 0.000009  min_lr: 0.000009  loss: 3.5595 (3.5700)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.0368 (1.0368)  acc1: 81.5104 (81.5104)  acc5: 96.0938 (96.0938)  time: 8.5515  data: 8.4328  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3574 (1.2770)  acc1: 73.1771 (74.0800)  acc5: 91.1458 (91.9840)  time: 1.0359  data: 0.9371  max mem: 21002
Test: Total time: 0:00:09 (1.0678 s / it)
* Acc@1 73.992 Acc@5 92.184 loss 1.282
Accuracy of the model on the 50000 test images: 74.0%
Max accuracy: 74.03%
Epoch: [292]  [  0/312]  eta: 1:04:08  lr: 0.000009  min_lr: 0.000009  loss: 3.9522 (3.9522)  weight_decay: 0.0500 (0.0500)  time: 12.3363  data: 10.0083  max mem: 21002
Epoch: [292]  [ 10/312]  eta: 0:08:46  lr: 0.000009  min_lr: 0.000009  loss: 3.7970 (3.6075)  weight_decay: 0.0500 (0.0500)  time: 1.7445  data: 1.0353  max mem: 21002
Epoch: [292]  [ 20/312]  eta: 0:05:14  lr: 0.000009  min_lr: 0.000009  loss: 3.5264 (3.5495)  weight_decay: 0.0500 (0.0500)  time: 0.5139  data: 0.0978  max mem: 21002
Epoch: [292]  [ 30/312]  eta: 0:03:52  lr: 0.000009  min_lr: 0.000009  loss: 3.1778 (3.4057)  weight_decay: 0.0500 (0.0500)  time: 0.3175  data: 0.0296  max mem: 21002
Epoch: [292]  [ 40/312]  eta: 0:03:09  lr: 0.000009  min_lr: 0.000009  loss: 3.3915 (3.4219)  weight_decay: 0.0500 (0.0500)  time: 0.2959  data: 0.0022  max mem: 21002
Epoch: [292]  [ 50/312]  eta: 0:02:56  lr: 0.000009  min_lr: 0.000009  loss: 3.5536 (3.4235)  weight_decay: 0.0500 (0.0500)  time: 0.4376  data: 0.0712  max mem: 21002
Epoch: [292]  [ 60/312]  eta: 0:02:38  lr: 0.000009  min_lr: 0.000009  loss: 3.5522 (3.4391)  weight_decay: 0.0500 (0.0500)  time: 0.4962  data: 0.0846  max mem: 21002
Epoch: [292]  [ 70/312]  eta: 0:02:29  lr: 0.000009  min_lr: 0.000009  loss: 3.5522 (3.4465)  weight_decay: 0.0500 (0.0500)  time: 0.4819  data: 0.1160  max mem: 21002
Epoch: [292]  [ 80/312]  eta: 0:02:24  lr: 0.000009  min_lr: 0.000009  loss: 3.5571 (3.4660)  weight_decay: 0.0500 (0.0500)  time: 0.5918  data: 0.1946  max mem: 21002
Epoch: [292]  [ 90/312]  eta: 0:02:14  lr: 0.000008  min_lr: 0.000008  loss: 3.7906 (3.5195)  weight_decay: 0.0500 (0.0500)  time: 0.5702  data: 0.1407  max mem: 21002
Epoch: [292]  [100/312]  eta: 0:02:07  lr: 0.000008  min_lr: 0.000008  loss: 3.8684 (3.5545)  weight_decay: 0.0500 (0.0500)  time: 0.5179  data: 0.1225  max mem: 21002
Epoch: [292]  [110/312]  eta: 0:01:58  lr: 0.000008  min_lr: 0.000008  loss: 3.7088 (3.5365)  weight_decay: 0.0500 (0.0500)  time: 0.4907  data: 0.1175  max mem: 21002
Epoch: [292]  [120/312]  eta: 0:01:53  lr: 0.000008  min_lr: 0.000008  loss: 3.7554 (3.5636)  weight_decay: 0.0500 (0.0500)  time: 0.5286  data: 0.1358  max mem: 21002
Epoch: [292]  [130/312]  eta: 0:01:48  lr: 0.000008  min_lr: 0.000008  loss: 3.8282 (3.5753)  weight_decay: 0.0500 (0.0500)  time: 0.6467  data: 0.1662  max mem: 21002
Epoch: [292]  [140/312]  eta: 0:01:39  lr: 0.000008  min_lr: 0.000008  loss: 3.8304 (3.5831)  weight_decay: 0.0500 (0.0500)  time: 0.5370  data: 0.1139  max mem: 21002
Epoch: [292]  [150/312]  eta: 0:01:34  lr: 0.000008  min_lr: 0.000008  loss: 3.7549 (3.5901)  weight_decay: 0.0500 (0.0500)  time: 0.4965  data: 0.1212  max mem: 21002
Epoch: [292]  [160/312]  eta: 0:01:28  lr: 0.000008  min_lr: 0.000008  loss: 3.7204 (3.5895)  weight_decay: 0.0500 (0.0500)  time: 0.5791  data: 0.1735  max mem: 21002
Epoch: [292]  [170/312]  eta: 0:01:22  lr: 0.000008  min_lr: 0.000008  loss: 3.7204 (3.5854)  weight_decay: 0.0500 (0.0500)  time: 0.5665  data: 0.0978  max mem: 21002
Epoch: [292]  [180/312]  eta: 0:01:17  lr: 0.000008  min_lr: 0.000008  loss: 3.6886 (3.5923)  weight_decay: 0.0500 (0.0500)  time: 0.6331  data: 0.0588  max mem: 21002
Epoch: [292]  [190/312]  eta: 0:01:10  lr: 0.000008  min_lr: 0.000008  loss: 3.7313 (3.5955)  weight_decay: 0.0500 (0.0500)  time: 0.5437  data: 0.0596  max mem: 21002
Epoch: [292]  [200/312]  eta: 0:01:04  lr: 0.000008  min_lr: 0.000008  loss: 3.7313 (3.5961)  weight_decay: 0.0500 (0.0500)  time: 0.5101  data: 0.0558  max mem: 21002
Epoch: [292]  [210/312]  eta: 0:00:59  lr: 0.000008  min_lr: 0.000008  loss: 3.7852 (3.6025)  weight_decay: 0.0500 (0.0500)  time: 0.6259  data: 0.1160  max mem: 21002
Epoch: [292]  [220/312]  eta: 0:00:52  lr: 0.000008  min_lr: 0.000008  loss: 3.7731 (3.6006)  weight_decay: 0.0500 (0.0500)  time: 0.5005  data: 0.0657  max mem: 21002
Epoch: [292]  [230/312]  eta: 0:00:47  lr: 0.000008  min_lr: 0.000008  loss: 3.8801 (3.6138)  weight_decay: 0.0500 (0.0500)  time: 0.4878  data: 0.0443  max mem: 21002
Epoch: [292]  [240/312]  eta: 0:00:41  lr: 0.000008  min_lr: 0.000008  loss: 3.8801 (3.6122)  weight_decay: 0.0500 (0.0500)  time: 0.5477  data: 0.1143  max mem: 21002
Epoch: [292]  [250/312]  eta: 0:00:35  lr: 0.000008  min_lr: 0.000008  loss: 3.6914 (3.6117)  weight_decay: 0.0500 (0.0500)  time: 0.5109  data: 0.0709  max mem: 21002
Epoch: [292]  [260/312]  eta: 0:00:29  lr: 0.000007  min_lr: 0.000007  loss: 3.5633 (3.5921)  weight_decay: 0.0500 (0.0500)  time: 0.6072  data: 0.0934  max mem: 21002
Epoch: [292]  [270/312]  eta: 0:00:23  lr: 0.000007  min_lr: 0.000007  loss: 3.2941 (3.5861)  weight_decay: 0.0500 (0.0500)  time: 0.4942  data: 0.0935  max mem: 21002
Epoch: [292]  [280/312]  eta: 0:00:18  lr: 0.000007  min_lr: 0.000007  loss: 3.4724 (3.5879)  weight_decay: 0.0500 (0.0500)  time: 0.4987  data: 0.1067  max mem: 21002
Epoch: [292]  [290/312]  eta: 0:00:12  lr: 0.000007  min_lr: 0.000007  loss: 3.6236 (3.5880)  weight_decay: 0.0500 (0.0500)  time: 0.6649  data: 0.2134  max mem: 21002
Epoch: [292]  [300/312]  eta: 0:00:06  lr: 0.000007  min_lr: 0.000007  loss: 3.6058 (3.5800)  weight_decay: 0.0500 (0.0500)  time: 0.4593  data: 0.1072  max mem: 21002
Epoch: [292]  [310/312]  eta: 0:00:01  lr: 0.000007  min_lr: 0.000007  loss: 3.4460 (3.5773)  weight_decay: 0.0500 (0.0500)  time: 0.2776  data: 0.0001  max mem: 21002
Epoch: [292]  [311/312]  eta: 0:00:00  lr: 0.000007  min_lr: 0.000007  loss: 3.5086 (3.5771)  weight_decay: 0.0500 (0.0500)  time: 0.2775  data: 0.0001  max mem: 21002
Epoch: [292] Total time: 0:02:52 (0.5544 s / it)
Averaged stats: lr: 0.000007  min_lr: 0.000007  loss: 3.5086 (3.5647)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 1.0198 (1.0198)  acc1: 81.7708 (81.7708)  acc5: 96.0938 (96.0938)  time: 8.9395  data: 8.8229  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3480 (1.2651)  acc1: 73.5677 (74.3040)  acc5: 91.7969 (92.1440)  time: 1.0790  data: 0.9804  max mem: 21002
Test: Total time: 0:00:09 (1.0944 s / it)
* Acc@1 73.972 Acc@5 92.162 loss 1.270
Accuracy of the model on the 50000 test images: 74.0%
Max accuracy: 74.03%
Epoch: [293]  [  0/312]  eta: 0:57:48  lr: 0.000007  min_lr: 0.000007  loss: 3.8553 (3.8553)  weight_decay: 0.0500 (0.0500)  time: 11.1161  data: 10.6059  max mem: 21002
Epoch: [293]  [ 10/312]  eta: 0:07:55  lr: 0.000007  min_lr: 0.000007  loss: 3.8553 (3.5658)  weight_decay: 0.0500 (0.0500)  time: 1.5751  data: 0.9648  max mem: 21002
Epoch: [293]  [ 20/312]  eta: 0:04:58  lr: 0.000007  min_lr: 0.000007  loss: 3.7460 (3.5477)  weight_decay: 0.0500 (0.0500)  time: 0.5170  data: 0.0574  max mem: 21002
Epoch: [293]  [ 30/312]  eta: 0:03:41  lr: 0.000007  min_lr: 0.000007  loss: 3.3289 (3.4072)  weight_decay: 0.0500 (0.0500)  time: 0.3515  data: 0.0574  max mem: 21002
Epoch: [293]  [ 40/312]  eta: 0:03:20  lr: 0.000007  min_lr: 0.000007  loss: 3.3289 (3.4341)  weight_decay: 0.0500 (0.0500)  time: 0.4346  data: 0.0966  max mem: 21002
Epoch: [293]  [ 50/312]  eta: 0:03:06  lr: 0.000007  min_lr: 0.000007  loss: 3.7841 (3.5245)  weight_decay: 0.0500 (0.0500)  time: 0.5964  data: 0.1990  max mem: 21002
Epoch: [293]  [ 60/312]  eta: 0:02:41  lr: 0.000007  min_lr: 0.000007  loss: 3.5754 (3.4861)  weight_decay: 0.0500 (0.0500)  time: 0.4517  data: 0.1031  max mem: 21002
Epoch: [293]  [ 70/312]  eta: 0:02:39  lr: 0.000007  min_lr: 0.000007  loss: 3.4986 (3.4741)  weight_decay: 0.0500 (0.0500)  time: 0.5224  data: 0.1456  max mem: 21002
Epoch: [293]  [ 80/312]  eta: 0:02:26  lr: 0.000007  min_lr: 0.000007  loss: 3.5422 (3.4866)  weight_decay: 0.0500 (0.0500)  time: 0.6054  data: 0.1813  max mem: 21002
Epoch: [293]  [ 90/312]  eta: 0:02:18  lr: 0.000007  min_lr: 0.000007  loss: 3.3186 (3.4656)  weight_decay: 0.0500 (0.0500)  time: 0.5032  data: 0.1173  max mem: 21002
Epoch: [293]  [100/312]  eta: 0:02:12  lr: 0.000007  min_lr: 0.000007  loss: 3.3497 (3.4623)  weight_decay: 0.0500 (0.0500)  time: 0.5939  data: 0.1775  max mem: 21002
Epoch: [293]  [110/312]  eta: 0:02:00  lr: 0.000007  min_lr: 0.000007  loss: 3.8794 (3.5086)  weight_decay: 0.0500 (0.0500)  time: 0.4634  data: 0.0966  max mem: 21002
Epoch: [293]  [120/312]  eta: 0:01:55  lr: 0.000007  min_lr: 0.000007  loss: 3.8425 (3.5148)  weight_decay: 0.0500 (0.0500)  time: 0.4908  data: 0.1129  max mem: 21002
Epoch: [293]  [130/312]  eta: 0:01:50  lr: 0.000006  min_lr: 0.000006  loss: 3.7169 (3.5204)  weight_decay: 0.0500 (0.0500)  time: 0.6785  data: 0.1748  max mem: 21002
Epoch: [293]  [140/312]  eta: 0:01:41  lr: 0.000006  min_lr: 0.000006  loss: 3.7850 (3.5265)  weight_decay: 0.0500 (0.0500)  time: 0.4919  data: 0.0766  max mem: 21002
Epoch: [293]  [150/312]  eta: 0:01:36  lr: 0.000006  min_lr: 0.000006  loss: 3.8102 (3.5350)  weight_decay: 0.0500 (0.0500)  time: 0.5094  data: 0.1175  max mem: 21002
Epoch: [293]  [160/312]  eta: 0:01:28  lr: 0.000006  min_lr: 0.000006  loss: 3.8146 (3.5474)  weight_decay: 0.0500 (0.0500)  time: 0.5568  data: 0.1033  max mem: 21002
Epoch: [293]  [170/312]  eta: 0:01:22  lr: 0.000006  min_lr: 0.000006  loss: 3.8657 (3.5664)  weight_decay: 0.0500 (0.0500)  time: 0.5049  data: 0.0860  max mem: 21002
Epoch: [293]  [180/312]  eta: 0:01:16  lr: 0.000006  min_lr: 0.000006  loss: 3.9007 (3.5770)  weight_decay: 0.0500 (0.0500)  time: 0.5747  data: 0.0903  max mem: 21002
Epoch: [293]  [190/312]  eta: 0:01:10  lr: 0.000006  min_lr: 0.000006  loss: 3.7274 (3.5740)  weight_decay: 0.0500 (0.0500)  time: 0.4850  data: 0.0681  max mem: 21002
Epoch: [293]  [200/312]  eta: 0:01:05  lr: 0.000006  min_lr: 0.000006  loss: 3.6260 (3.5811)  weight_decay: 0.0500 (0.0500)  time: 0.5630  data: 0.1200  max mem: 21002
Epoch: [293]  [210/312]  eta: 0:00:59  lr: 0.000006  min_lr: 0.000006  loss: 3.6260 (3.5677)  weight_decay: 0.0500 (0.0500)  time: 0.6462  data: 0.0569  max mem: 21002
Epoch: [293]  [220/312]  eta: 0:00:52  lr: 0.000006  min_lr: 0.000006  loss: 3.3751 (3.5565)  weight_decay: 0.0500 (0.0500)  time: 0.4768  data: 0.0417  max mem: 21002
Epoch: [293]  [230/312]  eta: 0:00:47  lr: 0.000006  min_lr: 0.000006  loss: 3.4415 (3.5545)  weight_decay: 0.0500 (0.0500)  time: 0.5649  data: 0.0950  max mem: 21002
Epoch: [293]  [240/312]  eta: 0:00:41  lr: 0.000006  min_lr: 0.000006  loss: 3.6661 (3.5609)  weight_decay: 0.0500 (0.0500)  time: 0.5517  data: 0.0541  max mem: 21002
Epoch: [293]  [250/312]  eta: 0:00:35  lr: 0.000006  min_lr: 0.000006  loss: 3.5969 (3.5507)  weight_decay: 0.0500 (0.0500)  time: 0.4733  data: 0.0198  max mem: 21002
Epoch: [293]  [260/312]  eta: 0:00:29  lr: 0.000006  min_lr: 0.000006  loss: 2.9634 (3.5319)  weight_decay: 0.0500 (0.0500)  time: 0.6309  data: 0.0198  max mem: 21002
Epoch: [293]  [270/312]  eta: 0:00:23  lr: 0.000006  min_lr: 0.000006  loss: 3.2786 (3.5382)  weight_decay: 0.0500 (0.0500)  time: 0.5278  data: 0.0540  max mem: 21002
Epoch: [293]  [280/312]  eta: 0:00:18  lr: 0.000006  min_lr: 0.000006  loss: 3.5796 (3.5407)  weight_decay: 0.0500 (0.0500)  time: 0.5595  data: 0.1146  max mem: 21002
Epoch: [293]  [290/312]  eta: 0:00:12  lr: 0.000006  min_lr: 0.000006  loss: 3.5796 (3.5345)  weight_decay: 0.0500 (0.0500)  time: 0.5675  data: 0.0612  max mem: 21002
Epoch: [293]  [300/312]  eta: 0:00:06  lr: 0.000006  min_lr: 0.000006  loss: 3.6009 (3.5317)  weight_decay: 0.0500 (0.0500)  time: 0.3484  data: 0.0036  max mem: 21002
Epoch: [293]  [310/312]  eta: 0:00:01  lr: 0.000006  min_lr: 0.000006  loss: 3.7019 (3.5363)  weight_decay: 0.0500 (0.0500)  time: 0.2801  data: 0.0035  max mem: 21002
Epoch: [293]  [311/312]  eta: 0:00:00  lr: 0.000006  min_lr: 0.000006  loss: 3.5543 (3.5360)  weight_decay: 0.0500 (0.0500)  time: 0.2800  data: 0.0035  max mem: 21002
Epoch: [293] Total time: 0:02:52 (0.5525 s / it)
Averaged stats: lr: 0.000006  min_lr: 0.000006  loss: 3.5543 (3.5548)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.0334 (1.0334)  acc1: 81.3802 (81.3802)  acc5: 96.0938 (96.0938)  time: 8.8317  data: 8.7130  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3635 (1.2798)  acc1: 73.3073 (74.1440)  acc5: 92.0573 (92.1280)  time: 1.0680  data: 0.9682  max mem: 21002
Test: Total time: 0:00:09 (1.0896 s / it)
* Acc@1 73.964 Acc@5 92.174 loss 1.284
Accuracy of the model on the 50000 test images: 74.0%
Max accuracy: 74.03%
Epoch: [294]  [  0/312]  eta: 1:03:06  lr: 0.000006  min_lr: 0.000006  loss: 2.9238 (2.9238)  weight_decay: 0.0500 (0.0500)  time: 12.1369  data: 11.8521  max mem: 21002
Epoch: [294]  [ 10/312]  eta: 0:07:44  lr: 0.000005  min_lr: 0.000005  loss: 3.6570 (3.4880)  weight_decay: 0.0500 (0.0500)  time: 1.5366  data: 1.0780  max mem: 21002
Epoch: [294]  [ 20/312]  eta: 0:05:01  lr: 0.000005  min_lr: 0.000005  loss: 3.5594 (3.4810)  weight_decay: 0.0500 (0.0500)  time: 0.4782  data: 0.0738  max mem: 21002
Epoch: [294]  [ 30/312]  eta: 0:03:43  lr: 0.000005  min_lr: 0.000005  loss: 3.5661 (3.5365)  weight_decay: 0.0500 (0.0500)  time: 0.3846  data: 0.0739  max mem: 21002
Epoch: [294]  [ 40/312]  eta: 0:03:25  lr: 0.000005  min_lr: 0.000005  loss: 3.7246 (3.5439)  weight_decay: 0.0500 (0.0500)  time: 0.4636  data: 0.1300  max mem: 21002
Epoch: [294]  [ 50/312]  eta: 0:03:11  lr: 0.000005  min_lr: 0.000005  loss: 3.6617 (3.5340)  weight_decay: 0.0500 (0.0500)  time: 0.6306  data: 0.2730  max mem: 21002
Epoch: [294]  [ 60/312]  eta: 0:02:48  lr: 0.000005  min_lr: 0.000005  loss: 3.6039 (3.5538)  weight_decay: 0.0500 (0.0500)  time: 0.4927  data: 0.1439  max mem: 21002
Epoch: [294]  [ 70/312]  eta: 0:02:41  lr: 0.000005  min_lr: 0.000005  loss: 3.6039 (3.5695)  weight_decay: 0.0500 (0.0500)  time: 0.5038  data: 0.1251  max mem: 21002
Epoch: [294]  [ 80/312]  eta: 0:02:34  lr: 0.000005  min_lr: 0.000005  loss: 3.3416 (3.5378)  weight_decay: 0.0500 (0.0500)  time: 0.6488  data: 0.2573  max mem: 21002
Epoch: [294]  [ 90/312]  eta: 0:02:18  lr: 0.000005  min_lr: 0.000005  loss: 3.5363 (3.5622)  weight_decay: 0.0500 (0.0500)  time: 0.4817  data: 0.1330  max mem: 21002
Epoch: [294]  [100/312]  eta: 0:02:13  lr: 0.000005  min_lr: 0.000005  loss: 3.6251 (3.5353)  weight_decay: 0.0500 (0.0500)  time: 0.4841  data: 0.1140  max mem: 21002
Epoch: [294]  [110/312]  eta: 0:02:00  lr: 0.000005  min_lr: 0.000005  loss: 3.5229 (3.5391)  weight_decay: 0.0500 (0.0500)  time: 0.4700  data: 0.1140  max mem: 21002
Epoch: [294]  [120/312]  eta: 0:01:56  lr: 0.000005  min_lr: 0.000005  loss: 3.5578 (3.5360)  weight_decay: 0.0500 (0.0500)  time: 0.4905  data: 0.0481  max mem: 21002
Epoch: [294]  [130/312]  eta: 0:01:50  lr: 0.000005  min_lr: 0.000005  loss: 3.7449 (3.5547)  weight_decay: 0.0500 (0.0500)  time: 0.6785  data: 0.1084  max mem: 21002
Epoch: [294]  [140/312]  eta: 0:01:41  lr: 0.000005  min_lr: 0.000005  loss: 3.6748 (3.5367)  weight_decay: 0.0500 (0.0500)  time: 0.4760  data: 0.0614  max mem: 21002
Epoch: [294]  [150/312]  eta: 0:01:36  lr: 0.000005  min_lr: 0.000005  loss: 3.4802 (3.5474)  weight_decay: 0.0500 (0.0500)  time: 0.5229  data: 0.0593  max mem: 21002
Epoch: [294]  [160/312]  eta: 0:01:30  lr: 0.000005  min_lr: 0.000005  loss: 3.7432 (3.5447)  weight_decay: 0.0500 (0.0500)  time: 0.6560  data: 0.0588  max mem: 21002
Epoch: [294]  [170/312]  eta: 0:01:22  lr: 0.000005  min_lr: 0.000005  loss: 3.3766 (3.5298)  weight_decay: 0.0500 (0.0500)  time: 0.4479  data: 0.0168  max mem: 21002
Epoch: [294]  [180/312]  eta: 0:01:17  lr: 0.000005  min_lr: 0.000005  loss: 3.6482 (3.5465)  weight_decay: 0.0500 (0.0500)  time: 0.5400  data: 0.0618  max mem: 21002
Epoch: [294]  [190/312]  eta: 0:01:10  lr: 0.000005  min_lr: 0.000005  loss: 3.6226 (3.5237)  weight_decay: 0.0500 (0.0500)  time: 0.5184  data: 0.0469  max mem: 21002
Epoch: [294]  [200/312]  eta: 0:01:04  lr: 0.000005  min_lr: 0.000005  loss: 3.2582 (3.5226)  weight_decay: 0.0500 (0.0500)  time: 0.4564  data: 0.0215  max mem: 21002
Epoch: [294]  [210/312]  eta: 0:00:59  lr: 0.000005  min_lr: 0.000005  loss: 3.3797 (3.5182)  weight_decay: 0.0500 (0.0500)  time: 0.6562  data: 0.0322  max mem: 21002
Epoch: [294]  [220/312]  eta: 0:00:52  lr: 0.000005  min_lr: 0.000005  loss: 3.1518 (3.4980)  weight_decay: 0.0500 (0.0500)  time: 0.4907  data: 0.0125  max mem: 21002
Epoch: [294]  [230/312]  eta: 0:00:46  lr: 0.000004  min_lr: 0.000004  loss: 3.5444 (3.5113)  weight_decay: 0.0500 (0.0500)  time: 0.4735  data: 0.0405  max mem: 21002
Epoch: [294]  [240/312]  eta: 0:00:41  lr: 0.000004  min_lr: 0.000004  loss: 3.8480 (3.5103)  weight_decay: 0.0500 (0.0500)  time: 0.6625  data: 0.0863  max mem: 21002
Epoch: [294]  [250/312]  eta: 0:00:35  lr: 0.000004  min_lr: 0.000004  loss: 3.7232 (3.5087)  weight_decay: 0.0500 (0.0500)  time: 0.4761  data: 0.0464  max mem: 21002
Epoch: [294]  [260/312]  eta: 0:00:29  lr: 0.000004  min_lr: 0.000004  loss: 3.4363 (3.5099)  weight_decay: 0.0500 (0.0500)  time: 0.5094  data: 0.0352  max mem: 21002
Epoch: [294]  [270/312]  eta: 0:00:23  lr: 0.000004  min_lr: 0.000004  loss: 3.7263 (3.5161)  weight_decay: 0.0500 (0.0500)  time: 0.5098  data: 0.0352  max mem: 21002
Epoch: [294]  [280/312]  eta: 0:00:18  lr: 0.000004  min_lr: 0.000004  loss: 3.7263 (3.5219)  weight_decay: 0.0500 (0.0500)  time: 0.4645  data: 0.0246  max mem: 21002
Epoch: [294]  [290/312]  eta: 0:00:12  lr: 0.000004  min_lr: 0.000004  loss: 3.5250 (3.5192)  weight_decay: 0.0500 (0.0500)  time: 0.6211  data: 0.0570  max mem: 21002
Epoch: [294]  [300/312]  eta: 0:00:06  lr: 0.000004  min_lr: 0.000004  loss: 3.5545 (3.5261)  weight_decay: 0.0500 (0.0500)  time: 0.4389  data: 0.0327  max mem: 21002
Epoch: [294]  [310/312]  eta: 0:00:01  lr: 0.000004  min_lr: 0.000004  loss: 3.7591 (3.5291)  weight_decay: 0.0500 (0.0500)  time: 0.2772  data: 0.0001  max mem: 21002
Epoch: [294]  [311/312]  eta: 0:00:00  lr: 0.000004  min_lr: 0.000004  loss: 3.7591 (3.5298)  weight_decay: 0.0500 (0.0500)  time: 0.2772  data: 0.0001  max mem: 21002
Epoch: [294] Total time: 0:02:51 (0.5494 s / it)
Averaged stats: lr: 0.000004  min_lr: 0.000004  loss: 3.7591 (3.5533)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.0984 (1.0984)  acc1: 81.5104 (81.5104)  acc5: 95.8333 (95.8333)  time: 8.3842  data: 8.2653  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4139 (1.3307)  acc1: 73.6979 (74.0000)  acc5: 92.1875 (92.1440)  time: 1.0194  data: 0.9185  max mem: 21002
Test: Total time: 0:00:09 (1.0532 s / it)
* Acc@1 73.824 Acc@5 92.126 loss 1.337
Accuracy of the model on the 50000 test images: 73.8%
Max accuracy: 74.03%
Epoch: [295]  [  0/312]  eta: 1:00:05  lr: 0.000004  min_lr: 0.000004  loss: 3.8172 (3.8172)  weight_decay: 0.0500 (0.0500)  time: 11.5551  data: 9.6874  max mem: 21002
Epoch: [295]  [ 10/312]  eta: 0:07:59  lr: 0.000004  min_lr: 0.000004  loss: 3.8236 (3.6820)  weight_decay: 0.0500 (0.0500)  time: 1.5884  data: 0.9690  max mem: 21002
Epoch: [295]  [ 20/312]  eta: 0:05:15  lr: 0.000004  min_lr: 0.000004  loss: 3.7258 (3.4745)  weight_decay: 0.0500 (0.0500)  time: 0.5552  data: 0.1205  max mem: 21002
Epoch: [295]  [ 30/312]  eta: 0:03:53  lr: 0.000004  min_lr: 0.000004  loss: 3.3659 (3.4503)  weight_decay: 0.0500 (0.0500)  time: 0.4076  data: 0.0723  max mem: 21002
Epoch: [295]  [ 40/312]  eta: 0:03:22  lr: 0.000004  min_lr: 0.000004  loss: 3.7719 (3.5287)  weight_decay: 0.0500 (0.0500)  time: 0.3945  data: 0.0049  max mem: 21002
Epoch: [295]  [ 50/312]  eta: 0:03:16  lr: 0.000004  min_lr: 0.000004  loss: 3.8947 (3.5868)  weight_decay: 0.0500 (0.0500)  time: 0.6331  data: 0.0564  max mem: 21002
Epoch: [295]  [ 60/312]  eta: 0:02:51  lr: 0.000004  min_lr: 0.000004  loss: 3.8739 (3.5996)  weight_decay: 0.0500 (0.0500)  time: 0.5448  data: 0.0673  max mem: 21002
Epoch: [295]  [ 70/312]  eta: 0:02:40  lr: 0.000004  min_lr: 0.000004  loss: 3.6508 (3.5925)  weight_decay: 0.0500 (0.0500)  time: 0.4468  data: 0.0473  max mem: 21002
Epoch: [295]  [ 80/312]  eta: 0:02:29  lr: 0.000004  min_lr: 0.000004  loss: 3.5718 (3.5971)  weight_decay: 0.0500 (0.0500)  time: 0.5320  data: 0.0324  max mem: 21002
Epoch: [295]  [ 90/312]  eta: 0:02:17  lr: 0.000004  min_lr: 0.000004  loss: 3.6456 (3.6095)  weight_decay: 0.0500 (0.0500)  time: 0.4632  data: 0.0281  max mem: 21002
Epoch: [295]  [100/312]  eta: 0:02:12  lr: 0.000004  min_lr: 0.000004  loss: 3.6020 (3.5823)  weight_decay: 0.0500 (0.0500)  time: 0.5610  data: 0.0873  max mem: 21002
Epoch: [295]  [110/312]  eta: 0:02:00  lr: 0.000004  min_lr: 0.000004  loss: 3.4164 (3.5681)  weight_decay: 0.0500 (0.0500)  time: 0.4840  data: 0.0600  max mem: 21002
Epoch: [295]  [120/312]  eta: 0:01:56  lr: 0.000004  min_lr: 0.000004  loss: 3.4343 (3.5683)  weight_decay: 0.0500 (0.0500)  time: 0.5127  data: 0.0278  max mem: 21002
Epoch: [295]  [130/312]  eta: 0:01:50  lr: 0.000004  min_lr: 0.000004  loss: 3.7788 (3.5734)  weight_decay: 0.0500 (0.0500)  time: 0.6745  data: 0.0642  max mem: 21002
Epoch: [295]  [140/312]  eta: 0:01:41  lr: 0.000004  min_lr: 0.000004  loss: 3.4865 (3.5636)  weight_decay: 0.0500 (0.0500)  time: 0.4625  data: 0.0371  max mem: 21002
Epoch: [295]  [150/312]  eta: 0:01:36  lr: 0.000004  min_lr: 0.000004  loss: 3.4364 (3.5492)  weight_decay: 0.0500 (0.0500)  time: 0.4968  data: 0.0469  max mem: 21002
Epoch: [295]  [160/312]  eta: 0:01:29  lr: 0.000004  min_lr: 0.000004  loss: 3.5221 (3.5623)  weight_decay: 0.0500 (0.0500)  time: 0.6058  data: 0.0471  max mem: 21002
Epoch: [295]  [170/312]  eta: 0:01:22  lr: 0.000003  min_lr: 0.000003  loss: 3.8392 (3.5776)  weight_decay: 0.0500 (0.0500)  time: 0.4898  data: 0.0222  max mem: 21002
Epoch: [295]  [180/312]  eta: 0:01:17  lr: 0.000003  min_lr: 0.000003  loss: 3.7260 (3.5771)  weight_decay: 0.0500 (0.0500)  time: 0.5733  data: 0.0593  max mem: 21002
Epoch: [295]  [190/312]  eta: 0:01:09  lr: 0.000003  min_lr: 0.000003  loss: 3.5411 (3.5640)  weight_decay: 0.0500 (0.0500)  time: 0.4976  data: 0.0380  max mem: 21002
Epoch: [295]  [200/312]  eta: 0:01:04  lr: 0.000003  min_lr: 0.000003  loss: 3.4938 (3.5696)  weight_decay: 0.0500 (0.0500)  time: 0.4854  data: 0.0265  max mem: 21002
Epoch: [295]  [210/312]  eta: 0:00:59  lr: 0.000003  min_lr: 0.000003  loss: 3.4938 (3.5655)  weight_decay: 0.0500 (0.0500)  time: 0.6865  data: 0.0316  max mem: 21002
Epoch: [295]  [220/312]  eta: 0:00:52  lr: 0.000003  min_lr: 0.000003  loss: 3.5609 (3.5659)  weight_decay: 0.0500 (0.0500)  time: 0.4949  data: 0.0059  max mem: 21002
Epoch: [295]  [230/312]  eta: 0:00:47  lr: 0.000003  min_lr: 0.000003  loss: 3.8153 (3.5695)  weight_decay: 0.0500 (0.0500)  time: 0.4753  data: 0.0274  max mem: 21002
Epoch: [295]  [240/312]  eta: 0:00:40  lr: 0.000003  min_lr: 0.000003  loss: 3.7624 (3.5640)  weight_decay: 0.0500 (0.0500)  time: 0.5555  data: 0.0273  max mem: 21002
Epoch: [295]  [250/312]  eta: 0:00:35  lr: 0.000003  min_lr: 0.000003  loss: 3.6167 (3.5601)  weight_decay: 0.0500 (0.0500)  time: 0.5207  data: 0.0470  max mem: 21002
Epoch: [295]  [260/312]  eta: 0:00:29  lr: 0.000003  min_lr: 0.000003  loss: 3.6088 (3.5605)  weight_decay: 0.0500 (0.0500)  time: 0.6183  data: 0.0742  max mem: 21002
Epoch: [295]  [270/312]  eta: 0:00:23  lr: 0.000003  min_lr: 0.000003  loss: 3.6830 (3.5642)  weight_decay: 0.0500 (0.0500)  time: 0.4689  data: 0.0279  max mem: 21002
Epoch: [295]  [280/312]  eta: 0:00:18  lr: 0.000003  min_lr: 0.000003  loss: 3.6325 (3.5568)  weight_decay: 0.0500 (0.0500)  time: 0.4978  data: 0.0094  max mem: 21002
Epoch: [295]  [290/312]  eta: 0:00:12  lr: 0.000003  min_lr: 0.000003  loss: 3.5479 (3.5559)  weight_decay: 0.0500 (0.0500)  time: 0.6374  data: 0.0204  max mem: 21002
Epoch: [295]  [300/312]  eta: 0:00:06  lr: 0.000003  min_lr: 0.000003  loss: 3.5390 (3.5452)  weight_decay: 0.0500 (0.0500)  time: 0.4261  data: 0.0115  max mem: 21002
Epoch: [295]  [310/312]  eta: 0:00:01  lr: 0.000003  min_lr: 0.000003  loss: 3.5843 (3.5417)  weight_decay: 0.0500 (0.0500)  time: 0.2807  data: 0.0001  max mem: 21002
Epoch: [295]  [311/312]  eta: 0:00:00  lr: 0.000003  min_lr: 0.000003  loss: 3.5843 (3.5422)  weight_decay: 0.0500 (0.0500)  time: 0.2806  data: 0.0001  max mem: 21002
Epoch: [295] Total time: 0:02:52 (0.5519 s / it)
Averaged stats: lr: 0.000003  min_lr: 0.000003  loss: 3.5843 (3.5659)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.0009 (1.0009)  acc1: 81.5104 (81.5104)  acc5: 96.2240 (96.2240)  time: 8.6212  data: 8.5031  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3263 (1.2433)  acc1: 73.5677 (74.1920)  acc5: 92.0573 (92.1280)  time: 1.0586  data: 0.9618  max mem: 21002
Test: Total time: 0:00:09 (1.0692 s / it)
* Acc@1 74.050 Acc@5 92.198 loss 1.247
Accuracy of the model on the 50000 test images: 74.1%
Max accuracy: 74.05%
Epoch: [296]  [  0/312]  eta: 0:59:05  lr: 0.000003  min_lr: 0.000003  loss: 3.8821 (3.8821)  weight_decay: 0.0500 (0.0500)  time: 11.3627  data: 10.0849  max mem: 21002
Epoch: [296]  [ 10/312]  eta: 0:07:35  lr: 0.000003  min_lr: 0.000003  loss: 3.2353 (3.3003)  weight_decay: 0.0500 (0.0500)  time: 1.5076  data: 1.0168  max mem: 21002
Epoch: [296]  [ 20/312]  eta: 0:05:04  lr: 0.000003  min_lr: 0.000003  loss: 3.2353 (3.3894)  weight_decay: 0.0500 (0.0500)  time: 0.5273  data: 0.1309  max mem: 21002
Epoch: [296]  [ 30/312]  eta: 0:03:45  lr: 0.000003  min_lr: 0.000003  loss: 3.4436 (3.4123)  weight_decay: 0.0500 (0.0500)  time: 0.4112  data: 0.0763  max mem: 21002
Epoch: [296]  [ 40/312]  eta: 0:03:27  lr: 0.000003  min_lr: 0.000003  loss: 3.6565 (3.4958)  weight_decay: 0.0500 (0.0500)  time: 0.4711  data: 0.1007  max mem: 21002
Epoch: [296]  [ 50/312]  eta: 0:03:10  lr: 0.000003  min_lr: 0.000003  loss: 3.6565 (3.4538)  weight_decay: 0.0500 (0.0500)  time: 0.6186  data: 0.2438  max mem: 21002
Epoch: [296]  [ 60/312]  eta: 0:02:46  lr: 0.000003  min_lr: 0.000003  loss: 3.7391 (3.5415)  weight_decay: 0.0500 (0.0500)  time: 0.4508  data: 0.1437  max mem: 21002
Epoch: [296]  [ 70/312]  eta: 0:02:41  lr: 0.000003  min_lr: 0.000003  loss: 3.9255 (3.5879)  weight_decay: 0.0500 (0.0500)  time: 0.5079  data: 0.1144  max mem: 21002
Epoch: [296]  [ 80/312]  eta: 0:02:32  lr: 0.000003  min_lr: 0.000003  loss: 3.8717 (3.6300)  weight_decay: 0.0500 (0.0500)  time: 0.6441  data: 0.2667  max mem: 21002
Epoch: [296]  [ 90/312]  eta: 0:02:18  lr: 0.000003  min_lr: 0.000003  loss: 3.8052 (3.6232)  weight_decay: 0.0500 (0.0500)  time: 0.4623  data: 0.1530  max mem: 21002
Epoch: [296]  [100/312]  eta: 0:02:14  lr: 0.000003  min_lr: 0.000003  loss: 3.7054 (3.6281)  weight_decay: 0.0500 (0.0500)  time: 0.5337  data: 0.1341  max mem: 21002
Epoch: [296]  [110/312]  eta: 0:02:02  lr: 0.000003  min_lr: 0.000003  loss: 3.5479 (3.6051)  weight_decay: 0.0500 (0.0500)  time: 0.5407  data: 0.1340  max mem: 21002
Epoch: [296]  [120/312]  eta: 0:01:56  lr: 0.000003  min_lr: 0.000003  loss: 3.5121 (3.6090)  weight_decay: 0.0500 (0.0500)  time: 0.4735  data: 0.1036  max mem: 21002
Epoch: [296]  [130/312]  eta: 0:01:50  lr: 0.000003  min_lr: 0.000003  loss: 3.7258 (3.6174)  weight_decay: 0.0500 (0.0500)  time: 0.6120  data: 0.2417  max mem: 21002
Epoch: [296]  [140/312]  eta: 0:01:40  lr: 0.000003  min_lr: 0.000003  loss: 3.8684 (3.6279)  weight_decay: 0.0500 (0.0500)  time: 0.4662  data: 0.1389  max mem: 21002
Epoch: [296]  [150/312]  eta: 0:01:35  lr: 0.000003  min_lr: 0.000003  loss: 3.7649 (3.6335)  weight_decay: 0.0500 (0.0500)  time: 0.4897  data: 0.1940  max mem: 21002
Epoch: [296]  [160/312]  eta: 0:01:30  lr: 0.000003  min_lr: 0.000003  loss: 3.4988 (3.6114)  weight_decay: 0.0500 (0.0500)  time: 0.6526  data: 0.3668  max mem: 21002
Epoch: [296]  [170/312]  eta: 0:01:22  lr: 0.000003  min_lr: 0.000003  loss: 3.1205 (3.5968)  weight_decay: 0.0500 (0.0500)  time: 0.4712  data: 0.1863  max mem: 21002
Epoch: [296]  [180/312]  eta: 0:01:17  lr: 0.000002  min_lr: 0.000002  loss: 3.7406 (3.6044)  weight_decay: 0.0500 (0.0500)  time: 0.5001  data: 0.1904  max mem: 21002
Epoch: [296]  [190/312]  eta: 0:01:09  lr: 0.000002  min_lr: 0.000002  loss: 3.7049 (3.5978)  weight_decay: 0.0500 (0.0500)  time: 0.4884  data: 0.1776  max mem: 21002
Epoch: [296]  [200/312]  eta: 0:01:04  lr: 0.000002  min_lr: 0.000002  loss: 3.6633 (3.6019)  weight_decay: 0.0500 (0.0500)  time: 0.5005  data: 0.1601  max mem: 21002
Epoch: [296]  [210/312]  eta: 0:00:59  lr: 0.000002  min_lr: 0.000002  loss: 3.7214 (3.6033)  weight_decay: 0.0500 (0.0500)  time: 0.6812  data: 0.3057  max mem: 21002
Epoch: [296]  [220/312]  eta: 0:00:52  lr: 0.000002  min_lr: 0.000002  loss: 3.7166 (3.6028)  weight_decay: 0.0500 (0.0500)  time: 0.4665  data: 0.1466  max mem: 21002
Epoch: [296]  [230/312]  eta: 0:00:47  lr: 0.000002  min_lr: 0.000002  loss: 3.7166 (3.6095)  weight_decay: 0.0500 (0.0500)  time: 0.5222  data: 0.1498  max mem: 21002
Epoch: [296]  [240/312]  eta: 0:00:41  lr: 0.000002  min_lr: 0.000002  loss: 3.7798 (3.6134)  weight_decay: 0.0500 (0.0500)  time: 0.7135  data: 0.2346  max mem: 21002
Epoch: [296]  [250/312]  eta: 0:00:35  lr: 0.000002  min_lr: 0.000002  loss: 3.7607 (3.6069)  weight_decay: 0.0500 (0.0500)  time: 0.5019  data: 0.1120  max mem: 21002
Epoch: [296]  [260/312]  eta: 0:00:29  lr: 0.000002  min_lr: 0.000002  loss: 3.3480 (3.5970)  weight_decay: 0.0500 (0.0500)  time: 0.5331  data: 0.1605  max mem: 21002
Epoch: [296]  [270/312]  eta: 0:00:23  lr: 0.000002  min_lr: 0.000002  loss: 3.6691 (3.6029)  weight_decay: 0.0500 (0.0500)  time: 0.5099  data: 0.1342  max mem: 21002
Epoch: [296]  [280/312]  eta: 0:00:18  lr: 0.000002  min_lr: 0.000002  loss: 3.5849 (3.5960)  weight_decay: 0.0500 (0.0500)  time: 0.5213  data: 0.1071  max mem: 21002
Epoch: [296]  [290/312]  eta: 0:00:12  lr: 0.000002  min_lr: 0.000002  loss: 3.5849 (3.5980)  weight_decay: 0.0500 (0.0500)  time: 0.5931  data: 0.1300  max mem: 21002
Epoch: [296]  [300/312]  eta: 0:00:06  lr: 0.000002  min_lr: 0.000002  loss: 3.8254 (3.6036)  weight_decay: 0.0500 (0.0500)  time: 0.3555  data: 0.0233  max mem: 21002
Epoch: [296]  [310/312]  eta: 0:00:01  lr: 0.000002  min_lr: 0.000002  loss: 3.5230 (3.5980)  weight_decay: 0.0500 (0.0500)  time: 0.2772  data: 0.0001  max mem: 21002
Epoch: [296]  [311/312]  eta: 0:00:00  lr: 0.000002  min_lr: 0.000002  loss: 3.5230 (3.5995)  weight_decay: 0.0500 (0.0500)  time: 0.2771  data: 0.0001  max mem: 21002
Epoch: [296] Total time: 0:02:51 (0.5506 s / it)
Averaged stats: lr: 0.000002  min_lr: 0.000002  loss: 3.5230 (3.5614)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.0477 (1.0477)  acc1: 81.5104 (81.5104)  acc5: 95.9635 (95.9635)  time: 8.6990  data: 8.5802  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3736 (1.2861)  acc1: 73.0469 (73.9520)  acc5: 91.7969 (92.1600)  time: 1.0521  data: 0.9534  max mem: 21002
Test: Total time: 0:00:09 (1.0692 s / it)
* Acc@1 73.910 Acc@5 92.128 loss 1.292
Accuracy of the model on the 50000 test images: 73.9%
Max accuracy: 74.05%
Epoch: [297]  [  0/312]  eta: 0:52:54  lr: 0.000002  min_lr: 0.000002  loss: 3.7613 (3.7613)  weight_decay: 0.0500 (0.0500)  time: 10.1755  data: 9.2655  max mem: 21002
Epoch: [297]  [ 10/312]  eta: 0:07:54  lr: 0.000002  min_lr: 0.000002  loss: 3.8690 (3.8602)  weight_decay: 0.0500 (0.0500)  time: 1.5728  data: 1.0509  max mem: 21002
Epoch: [297]  [ 20/312]  eta: 0:05:15  lr: 0.000002  min_lr: 0.000002  loss: 3.8312 (3.8072)  weight_decay: 0.0500 (0.0500)  time: 0.6261  data: 0.1772  max mem: 21002
Epoch: [297]  [ 30/312]  eta: 0:03:53  lr: 0.000002  min_lr: 0.000002  loss: 3.8020 (3.7432)  weight_decay: 0.0500 (0.0500)  time: 0.4163  data: 0.0632  max mem: 21002
Epoch: [297]  [ 40/312]  eta: 0:03:26  lr: 0.000002  min_lr: 0.000002  loss: 3.2991 (3.6017)  weight_decay: 0.0500 (0.0500)  time: 0.4191  data: 0.0887  max mem: 21002
Epoch: [297]  [ 50/312]  eta: 0:03:09  lr: 0.000002  min_lr: 0.000002  loss: 3.2368 (3.5714)  weight_decay: 0.0500 (0.0500)  time: 0.5612  data: 0.1590  max mem: 21002
Epoch: [297]  [ 60/312]  eta: 0:02:50  lr: 0.000002  min_lr: 0.000002  loss: 3.6818 (3.5912)  weight_decay: 0.0500 (0.0500)  time: 0.5151  data: 0.1553  max mem: 21002
Epoch: [297]  [ 70/312]  eta: 0:02:45  lr: 0.000002  min_lr: 0.000002  loss: 3.5172 (3.5483)  weight_decay: 0.0500 (0.0500)  time: 0.5839  data: 0.1885  max mem: 21002
Epoch: [297]  [ 80/312]  eta: 0:02:31  lr: 0.000002  min_lr: 0.000002  loss: 3.2258 (3.5378)  weight_decay: 0.0500 (0.0500)  time: 0.5740  data: 0.1046  max mem: 21002
Epoch: [297]  [ 90/312]  eta: 0:02:22  lr: 0.000002  min_lr: 0.000002  loss: 3.5632 (3.5283)  weight_decay: 0.0500 (0.0500)  time: 0.4863  data: 0.0738  max mem: 21002
Epoch: [297]  [100/312]  eta: 0:02:14  lr: 0.000002  min_lr: 0.000002  loss: 3.6857 (3.5237)  weight_decay: 0.0500 (0.0500)  time: 0.5724  data: 0.1483  max mem: 21002
Epoch: [297]  [110/312]  eta: 0:02:02  lr: 0.000002  min_lr: 0.000002  loss: 3.6474 (3.5298)  weight_decay: 0.0500 (0.0500)  time: 0.4467  data: 0.0751  max mem: 21002
Epoch: [297]  [120/312]  eta: 0:01:58  lr: 0.000002  min_lr: 0.000002  loss: 3.4225 (3.5207)  weight_decay: 0.0500 (0.0500)  time: 0.5113  data: 0.1201  max mem: 21002
Epoch: [297]  [130/312]  eta: 0:01:50  lr: 0.000002  min_lr: 0.000002  loss: 3.7512 (3.5369)  weight_decay: 0.0500 (0.0500)  time: 0.6030  data: 0.1203  max mem: 21002
Epoch: [297]  [140/312]  eta: 0:01:42  lr: 0.000002  min_lr: 0.000002  loss: 3.6039 (3.5161)  weight_decay: 0.0500 (0.0500)  time: 0.4699  data: 0.0929  max mem: 21002
Epoch: [297]  [150/312]  eta: 0:01:37  lr: 0.000002  min_lr: 0.000002  loss: 3.5277 (3.5359)  weight_decay: 0.0500 (0.0500)  time: 0.6025  data: 0.1782  max mem: 21002
Epoch: [297]  [160/312]  eta: 0:01:30  lr: 0.000002  min_lr: 0.000002  loss: 3.4767 (3.5252)  weight_decay: 0.0500 (0.0500)  time: 0.6114  data: 0.0861  max mem: 21002
Epoch: [297]  [170/312]  eta: 0:01:23  lr: 0.000002  min_lr: 0.000002  loss: 3.6140 (3.5354)  weight_decay: 0.0500 (0.0500)  time: 0.4652  data: 0.0768  max mem: 21002
Epoch: [297]  [180/312]  eta: 0:01:18  lr: 0.000002  min_lr: 0.000002  loss: 3.8094 (3.5530)  weight_decay: 0.0500 (0.0500)  time: 0.5832  data: 0.1456  max mem: 21002
Epoch: [297]  [190/312]  eta: 0:01:10  lr: 0.000002  min_lr: 0.000002  loss: 3.9005 (3.5717)  weight_decay: 0.0500 (0.0500)  time: 0.5053  data: 0.0702  max mem: 21002
Epoch: [297]  [200/312]  eta: 0:01:05  lr: 0.000002  min_lr: 0.000002  loss: 3.8110 (3.5723)  weight_decay: 0.0500 (0.0500)  time: 0.4659  data: 0.0597  max mem: 21002
Epoch: [297]  [210/312]  eta: 0:00:59  lr: 0.000002  min_lr: 0.000002  loss: 3.7881 (3.5783)  weight_decay: 0.0500 (0.0500)  time: 0.6203  data: 0.0593  max mem: 21002
Epoch: [297]  [220/312]  eta: 0:00:52  lr: 0.000002  min_lr: 0.000002  loss: 3.8126 (3.5830)  weight_decay: 0.0500 (0.0500)  time: 0.4878  data: 0.0486  max mem: 21002
Epoch: [297]  [230/312]  eta: 0:00:47  lr: 0.000002  min_lr: 0.000002  loss: 3.8810 (3.5881)  weight_decay: 0.0500 (0.0500)  time: 0.5258  data: 0.1050  max mem: 21002
Epoch: [297]  [240/312]  eta: 0:00:41  lr: 0.000002  min_lr: 0.000002  loss: 3.7948 (3.5941)  weight_decay: 0.0500 (0.0500)  time: 0.6048  data: 0.0573  max mem: 21002
Epoch: [297]  [250/312]  eta: 0:00:35  lr: 0.000002  min_lr: 0.000002  loss: 3.7344 (3.5995)  weight_decay: 0.0500 (0.0500)  time: 0.4936  data: 0.0617  max mem: 21002
Epoch: [297]  [260/312]  eta: 0:00:29  lr: 0.000002  min_lr: 0.000002  loss: 3.8361 (3.6066)  weight_decay: 0.0500 (0.0500)  time: 0.5348  data: 0.1198  max mem: 21002
Epoch: [297]  [270/312]  eta: 0:00:23  lr: 0.000002  min_lr: 0.000002  loss: 3.8630 (3.6069)  weight_decay: 0.0500 (0.0500)  time: 0.4530  data: 0.0591  max mem: 21002
Epoch: [297]  [280/312]  eta: 0:00:18  lr: 0.000002  min_lr: 0.000002  loss: 3.5899 (3.6039)  weight_decay: 0.0500 (0.0500)  time: 0.4786  data: 0.0678  max mem: 21002
Epoch: [297]  [290/312]  eta: 0:00:12  lr: 0.000002  min_lr: 0.000002  loss: 3.5809 (3.6047)  weight_decay: 0.0500 (0.0500)  time: 0.5852  data: 0.0673  max mem: 21002
Epoch: [297]  [300/312]  eta: 0:00:06  lr: 0.000002  min_lr: 0.000002  loss: 3.7679 (3.6092)  weight_decay: 0.0500 (0.0500)  time: 0.4240  data: 0.0358  max mem: 21002
Epoch: [297]  [310/312]  eta: 0:00:01  lr: 0.000002  min_lr: 0.000002  loss: 3.8632 (3.6211)  weight_decay: 0.0500 (0.0500)  time: 0.3125  data: 0.0357  max mem: 21002
Epoch: [297]  [311/312]  eta: 0:00:00  lr: 0.000002  min_lr: 0.000002  loss: 3.8099 (3.6174)  weight_decay: 0.0500 (0.0500)  time: 0.2774  data: 0.0001  max mem: 21002
Epoch: [297] Total time: 0:02:51 (0.5512 s / it)
Averaged stats: lr: 0.000002  min_lr: 0.000002  loss: 3.8099 (3.5784)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:22  loss: 1.0953 (1.0953)  acc1: 81.1198 (81.1198)  acc5: 95.7031 (95.7031)  time: 9.1634  data: 9.0449  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4156 (1.3278)  acc1: 73.1771 (73.9680)  acc5: 91.9271 (92.1440)  time: 1.1038  data: 1.0050  max mem: 21002
Test: Total time: 0:00:10 (1.1241 s / it)
* Acc@1 73.836 Acc@5 92.088 loss 1.333
Accuracy of the model on the 50000 test images: 73.8%
Max accuracy: 74.05%
Epoch: [298]  [  0/312]  eta: 0:56:23  lr: 0.000002  min_lr: 0.000002  loss: 4.0898 (4.0898)  weight_decay: 0.0500 (0.0500)  time: 10.8443  data: 9.2186  max mem: 21002
Epoch: [298]  [ 10/312]  eta: 0:07:55  lr: 0.000001  min_lr: 0.000001  loss: 3.7131 (3.5169)  weight_decay: 0.0500 (0.0500)  time: 1.5741  data: 1.0239  max mem: 21002
Epoch: [298]  [ 20/312]  eta: 0:05:46  lr: 0.000001  min_lr: 0.000001  loss: 3.4223 (3.4342)  weight_decay: 0.0500 (0.0500)  time: 0.7030  data: 0.1207  max mem: 21002
Epoch: [298]  [ 30/312]  eta: 0:04:12  lr: 0.000001  min_lr: 0.000001  loss: 3.4223 (3.4388)  weight_decay: 0.0500 (0.0500)  time: 0.5219  data: 0.0188  max mem: 21002
Epoch: [298]  [ 40/312]  eta: 0:03:30  lr: 0.000001  min_lr: 0.000001  loss: 3.4026 (3.4376)  weight_decay: 0.0500 (0.0500)  time: 0.3424  data: 0.0009  max mem: 21002
Epoch: [298]  [ 50/312]  eta: 0:03:12  lr: 0.000001  min_lr: 0.000001  loss: 3.4026 (3.4533)  weight_decay: 0.0500 (0.0500)  time: 0.4820  data: 0.0012  max mem: 21002
Epoch: [298]  [ 60/312]  eta: 0:02:50  lr: 0.000001  min_lr: 0.000001  loss: 3.5819 (3.4894)  weight_decay: 0.0500 (0.0500)  time: 0.4700  data: 0.0013  max mem: 21002
Epoch: [298]  [ 70/312]  eta: 0:02:43  lr: 0.000001  min_lr: 0.000001  loss: 3.5819 (3.4786)  weight_decay: 0.0500 (0.0500)  time: 0.5277  data: 0.0012  max mem: 21002
Epoch: [298]  [ 80/312]  eta: 0:02:33  lr: 0.000001  min_lr: 0.000001  loss: 3.6976 (3.4837)  weight_decay: 0.0500 (0.0500)  time: 0.6162  data: 0.0086  max mem: 21002
Epoch: [298]  [ 90/312]  eta: 0:02:19  lr: 0.000001  min_lr: 0.000001  loss: 3.5426 (3.4753)  weight_decay: 0.0500 (0.0500)  time: 0.4664  data: 0.0330  max mem: 21002
Epoch: [298]  [100/312]  eta: 0:02:16  lr: 0.000001  min_lr: 0.000001  loss: 3.4021 (3.4545)  weight_decay: 0.0500 (0.0500)  time: 0.5786  data: 0.0317  max mem: 21002
Epoch: [298]  [110/312]  eta: 0:02:04  lr: 0.000001  min_lr: 0.000001  loss: 3.3417 (3.4558)  weight_decay: 0.0500 (0.0500)  time: 0.5435  data: 0.0188  max mem: 21002
Epoch: [298]  [120/312]  eta: 0:01:59  lr: 0.000001  min_lr: 0.000001  loss: 3.3992 (3.4527)  weight_decay: 0.0500 (0.0500)  time: 0.5007  data: 0.0268  max mem: 21002
Epoch: [298]  [130/312]  eta: 0:01:52  lr: 0.000001  min_lr: 0.000001  loss: 3.5241 (3.4614)  weight_decay: 0.0500 (0.0500)  time: 0.6457  data: 0.0423  max mem: 21002
Epoch: [298]  [140/312]  eta: 0:01:43  lr: 0.000001  min_lr: 0.000001  loss: 3.2887 (3.4322)  weight_decay: 0.0500 (0.0500)  time: 0.4837  data: 0.0626  max mem: 21002
Epoch: [298]  [150/312]  eta: 0:01:38  lr: 0.000001  min_lr: 0.000001  loss: 3.2887 (3.4380)  weight_decay: 0.0500 (0.0500)  time: 0.5240  data: 0.0627  max mem: 21002
Epoch: [298]  [160/312]  eta: 0:01:32  lr: 0.000001  min_lr: 0.000001  loss: 3.6859 (3.4464)  weight_decay: 0.0500 (0.0500)  time: 0.6562  data: 0.0839  max mem: 21002
Epoch: [298]  [170/312]  eta: 0:01:24  lr: 0.000001  min_lr: 0.000001  loss: 3.5320 (3.4571)  weight_decay: 0.0500 (0.0500)  time: 0.5193  data: 0.1059  max mem: 21002
Epoch: [298]  [180/312]  eta: 0:01:18  lr: 0.000001  min_lr: 0.000001  loss: 3.7494 (3.4699)  weight_decay: 0.0500 (0.0500)  time: 0.5188  data: 0.0994  max mem: 21002
Epoch: [298]  [190/312]  eta: 0:01:11  lr: 0.000001  min_lr: 0.000001  loss: 3.7174 (3.4738)  weight_decay: 0.0500 (0.0500)  time: 0.5036  data: 0.0921  max mem: 21002
Epoch: [298]  [200/312]  eta: 0:01:05  lr: 0.000001  min_lr: 0.000001  loss: 3.4998 (3.4665)  weight_decay: 0.0500 (0.0500)  time: 0.5095  data: 0.0838  max mem: 21002
Epoch: [298]  [210/312]  eta: 0:01:00  lr: 0.000001  min_lr: 0.000001  loss: 3.5820 (3.4790)  weight_decay: 0.0500 (0.0500)  time: 0.6250  data: 0.0635  max mem: 21002
Epoch: [298]  [220/312]  eta: 0:00:53  lr: 0.000001  min_lr: 0.000001  loss: 3.7315 (3.4903)  weight_decay: 0.0500 (0.0500)  time: 0.4591  data: 0.0368  max mem: 21002
Epoch: [298]  [230/312]  eta: 0:00:47  lr: 0.000001  min_lr: 0.000001  loss: 3.6946 (3.4962)  weight_decay: 0.0500 (0.0500)  time: 0.4903  data: 0.0522  max mem: 21002
Epoch: [298]  [240/312]  eta: 0:00:42  lr: 0.000001  min_lr: 0.000001  loss: 3.6828 (3.5000)  weight_decay: 0.0500 (0.0500)  time: 0.6749  data: 0.0755  max mem: 21002
Epoch: [298]  [250/312]  eta: 0:00:35  lr: 0.000001  min_lr: 0.000001  loss: 3.6651 (3.4970)  weight_decay: 0.0500 (0.0500)  time: 0.5150  data: 0.0709  max mem: 21002
Epoch: [298]  [260/312]  eta: 0:00:30  lr: 0.000001  min_lr: 0.000001  loss: 3.6026 (3.5041)  weight_decay: 0.0500 (0.0500)  time: 0.5108  data: 0.0690  max mem: 21002
Epoch: [298]  [270/312]  eta: 0:00:24  lr: 0.000001  min_lr: 0.000001  loss: 3.8380 (3.5092)  weight_decay: 0.0500 (0.0500)  time: 0.5172  data: 0.0724  max mem: 21002
Epoch: [298]  [280/312]  eta: 0:00:18  lr: 0.000001  min_lr: 0.000001  loss: 3.2066 (3.4961)  weight_decay: 0.0500 (0.0500)  time: 0.5325  data: 0.1020  max mem: 21002
Epoch: [298]  [290/312]  eta: 0:00:12  lr: 0.000001  min_lr: 0.000001  loss: 3.5426 (3.5047)  weight_decay: 0.0500 (0.0500)  time: 0.5817  data: 0.0764  max mem: 21002
Epoch: [298]  [300/312]  eta: 0:00:06  lr: 0.000001  min_lr: 0.000001  loss: 3.7770 (3.5039)  weight_decay: 0.0500 (0.0500)  time: 0.3692  data: 0.0107  max mem: 21002
Epoch: [298]  [310/312]  eta: 0:00:01  lr: 0.000001  min_lr: 0.000001  loss: 3.5827 (3.5089)  weight_decay: 0.0500 (0.0500)  time: 0.2772  data: 0.0001  max mem: 21002
Epoch: [298]  [311/312]  eta: 0:00:00  lr: 0.000001  min_lr: 0.000001  loss: 3.7395 (3.5114)  weight_decay: 0.0500 (0.0500)  time: 0.2771  data: 0.0001  max mem: 21002
Epoch: [298] Total time: 0:02:53 (0.5564 s / it)
Averaged stats: lr: 0.000001  min_lr: 0.000001  loss: 3.7395 (3.5578)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.0769 (1.0769)  acc1: 81.3802 (81.3802)  acc5: 95.7031 (95.7031)  time: 8.6371  data: 8.5182  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.4026 (1.3183)  acc1: 73.4375 (74.0320)  acc5: 91.7969 (92.0480)  time: 1.0447  data: 0.9465  max mem: 21002
Test: Total time: 0:00:09 (1.0632 s / it)
* Acc@1 73.888 Acc@5 92.148 loss 1.324
Accuracy of the model on the 50000 test images: 73.9%
Max accuracy: 74.05%
Epoch: [299]  [  0/312]  eta: 0:59:09  lr: 0.000001  min_lr: 0.000001  loss: 3.0004 (3.0004)  weight_decay: 0.0500 (0.0500)  time: 11.3768  data: 10.8890  max mem: 21002
Epoch: [299]  [ 10/312]  eta: 0:08:00  lr: 0.000001  min_lr: 0.000001  loss: 3.8005 (3.6724)  weight_decay: 0.0500 (0.0500)  time: 1.5925  data: 1.0454  max mem: 21002
Epoch: [299]  [ 20/312]  eta: 0:05:06  lr: 0.000001  min_lr: 0.000001  loss: 3.8582 (3.6376)  weight_decay: 0.0500 (0.0500)  time: 0.5328  data: 0.0799  max mem: 21002
Epoch: [299]  [ 30/312]  eta: 0:03:46  lr: 0.000001  min_lr: 0.000001  loss: 3.8132 (3.6684)  weight_decay: 0.0500 (0.0500)  time: 0.3700  data: 0.0497  max mem: 21002
Epoch: [299]  [ 40/312]  eta: 0:03:22  lr: 0.000001  min_lr: 0.000001  loss: 3.8942 (3.7470)  weight_decay: 0.0500 (0.0500)  time: 0.4239  data: 0.0584  max mem: 21002
Epoch: [299]  [ 50/312]  eta: 0:03:12  lr: 0.000001  min_lr: 0.000001  loss: 3.8842 (3.6877)  weight_decay: 0.0500 (0.0500)  time: 0.6285  data: 0.1468  max mem: 21002
Epoch: [299]  [ 60/312]  eta: 0:02:46  lr: 0.000001  min_lr: 0.000001  loss: 3.7932 (3.7373)  weight_decay: 0.0500 (0.0500)  time: 0.4936  data: 0.0891  max mem: 21002
Epoch: [299]  [ 70/312]  eta: 0:02:40  lr: 0.000001  min_lr: 0.000001  loss: 3.7932 (3.7087)  weight_decay: 0.0500 (0.0500)  time: 0.4846  data: 0.0493  max mem: 21002
Epoch: [299]  [ 80/312]  eta: 0:02:31  lr: 0.000001  min_lr: 0.000001  loss: 3.4209 (3.6391)  weight_decay: 0.0500 (0.0500)  time: 0.6262  data: 0.0562  max mem: 21002
Epoch: [299]  [ 90/312]  eta: 0:02:19  lr: 0.000001  min_lr: 0.000001  loss: 3.6796 (3.6503)  weight_decay: 0.0500 (0.0500)  time: 0.4961  data: 0.0499  max mem: 21002
Epoch: [299]  [100/312]  eta: 0:02:14  lr: 0.000001  min_lr: 0.000001  loss: 3.6796 (3.6086)  weight_decay: 0.0500 (0.0500)  time: 0.5568  data: 0.0791  max mem: 21002
Epoch: [299]  [110/312]  eta: 0:02:01  lr: 0.000001  min_lr: 0.000001  loss: 3.2383 (3.5930)  weight_decay: 0.0500 (0.0500)  time: 0.4899  data: 0.0368  max mem: 21002
Epoch: [299]  [120/312]  eta: 0:01:56  lr: 0.000001  min_lr: 0.000001  loss: 3.7672 (3.5781)  weight_decay: 0.0500 (0.0500)  time: 0.4799  data: 0.0416  max mem: 21002
Epoch: [299]  [130/312]  eta: 0:01:51  lr: 0.000001  min_lr: 0.000001  loss: 3.7756 (3.5836)  weight_decay: 0.0500 (0.0500)  time: 0.6755  data: 0.0688  max mem: 21002
Epoch: [299]  [140/312]  eta: 0:01:41  lr: 0.000001  min_lr: 0.000001  loss: 3.7587 (3.5804)  weight_decay: 0.0500 (0.0500)  time: 0.4819  data: 0.0278  max mem: 21002
Epoch: [299]  [150/312]  eta: 0:01:36  lr: 0.000001  min_lr: 0.000001  loss: 3.1561 (3.5482)  weight_decay: 0.0500 (0.0500)  time: 0.4735  data: 0.0099  max mem: 21002
Epoch: [299]  [160/312]  eta: 0:01:31  lr: 0.000001  min_lr: 0.000001  loss: 3.3749 (3.5597)  weight_decay: 0.0500 (0.0500)  time: 0.6555  data: 0.0249  max mem: 21002
Epoch: [299]  [170/312]  eta: 0:01:22  lr: 0.000001  min_lr: 0.000001  loss: 3.5732 (3.5473)  weight_decay: 0.0500 (0.0500)  time: 0.4936  data: 0.0428  max mem: 21002
Epoch: [299]  [180/312]  eta: 0:01:17  lr: 0.000001  min_lr: 0.000001  loss: 3.6476 (3.5580)  weight_decay: 0.0500 (0.0500)  time: 0.5249  data: 0.0554  max mem: 21002
Epoch: [299]  [190/312]  eta: 0:01:10  lr: 0.000001  min_lr: 0.000001  loss: 3.8616 (3.5721)  weight_decay: 0.0500 (0.0500)  time: 0.5017  data: 0.0284  max mem: 21002
Epoch: [299]  [200/312]  eta: 0:01:04  lr: 0.000001  min_lr: 0.000001  loss: 3.5716 (3.5404)  weight_decay: 0.0500 (0.0500)  time: 0.4597  data: 0.0464  max mem: 21002
Epoch: [299]  [210/312]  eta: 0:00:59  lr: 0.000001  min_lr: 0.000001  loss: 3.3210 (3.5394)  weight_decay: 0.0500 (0.0500)  time: 0.6739  data: 0.0959  max mem: 21002
Epoch: [299]  [220/312]  eta: 0:00:52  lr: 0.000001  min_lr: 0.000001  loss: 3.5893 (3.5469)  weight_decay: 0.0500 (0.0500)  time: 0.5034  data: 0.0504  max mem: 21002
Epoch: [299]  [230/312]  eta: 0:00:47  lr: 0.000001  min_lr: 0.000001  loss: 3.7683 (3.5575)  weight_decay: 0.0500 (0.0500)  time: 0.4643  data: 0.0169  max mem: 21002
Epoch: [299]  [240/312]  eta: 0:00:41  lr: 0.000001  min_lr: 0.000001  loss: 3.8533 (3.5569)  weight_decay: 0.0500 (0.0500)  time: 0.6286  data: 0.0169  max mem: 21002
Epoch: [299]  [250/312]  eta: 0:00:35  lr: 0.000001  min_lr: 0.000001  loss: 3.4302 (3.5541)  weight_decay: 0.0500 (0.0500)  time: 0.4645  data: 0.0125  max mem: 21002
Epoch: [299]  [260/312]  eta: 0:00:29  lr: 0.000001  min_lr: 0.000001  loss: 3.5556 (3.5596)  weight_decay: 0.0500 (0.0500)  time: 0.4981  data: 0.0589  max mem: 21002
Epoch: [299]  [270/312]  eta: 0:00:23  lr: 0.000001  min_lr: 0.000001  loss: 3.7115 (3.5601)  weight_decay: 0.0500 (0.0500)  time: 0.4865  data: 0.0471  max mem: 21002
Epoch: [299]  [280/312]  eta: 0:00:18  lr: 0.000001  min_lr: 0.000001  loss: 3.7115 (3.5619)  weight_decay: 0.0500 (0.0500)  time: 0.4928  data: 0.0422  max mem: 21002
Epoch: [299]  [290/312]  eta: 0:00:12  lr: 0.000001  min_lr: 0.000001  loss: 3.5895 (3.5520)  weight_decay: 0.0500 (0.0500)  time: 0.6231  data: 0.0558  max mem: 21002
Epoch: [299]  [300/312]  eta: 0:00:06  lr: 0.000001  min_lr: 0.000001  loss: 3.2974 (3.5457)  weight_decay: 0.0500 (0.0500)  time: 0.4134  data: 0.0140  max mem: 21002
Epoch: [299]  [310/312]  eta: 0:00:01  lr: 0.000001  min_lr: 0.000001  loss: 3.5698 (3.5485)  weight_decay: 0.0500 (0.0500)  time: 0.2775  data: 0.0001  max mem: 21002
Epoch: [299]  [311/312]  eta: 0:00:00  lr: 0.000001  min_lr: 0.000001  loss: 3.5698 (3.5468)  weight_decay: 0.0500 (0.0500)  time: 0.2774  data: 0.0001  max mem: 21002
Epoch: [299] Total time: 0:02:50 (0.5480 s / it)
Averaged stats: lr: 0.000001  min_lr: 0.000001  loss: 3.5698 (3.5541)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.0065 (1.0065)  acc1: 81.6406 (81.6406)  acc5: 95.9635 (95.9635)  time: 8.5887  data: 8.4703  max mem: 21002
Test:  [8/9]  eta: 0:00:01  loss: 1.3224 (1.2430)  acc1: 73.1771 (74.1760)  acc5: 91.7969 (92.1920)  time: 1.0741  data: 0.9752  max mem: 21002
Test: Total time: 0:00:09 (1.1016 s / it)
* Acc@1 74.020 Acc@5 92.228 loss 1.248
Accuracy of the model on the 50000 test images: 74.0%
Max accuracy: 74.05%
Training time 5:51:10