| { | |
| "best_metric": 0.35714285714285715, | |
| "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-75", | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 9.760223388671875, | |
| "learning_rate": 4.850746268656717e-05, | |
| "loss": 2.2528, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 10.098638534545898, | |
| "learning_rate": 4.104477611940299e-05, | |
| "loss": 2.2017, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 9.566693305969238, | |
| "learning_rate": 3.358208955223881e-05, | |
| "loss": 2.106, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 9.71388053894043, | |
| "learning_rate": 2.6119402985074626e-05, | |
| "loss": 2.0158, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 13.246223449707031, | |
| "learning_rate": 1.865671641791045e-05, | |
| "loss": 2.0417, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 10.271443367004395, | |
| "learning_rate": 1.119402985074627e-05, | |
| "loss": 1.8307, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 16.004535675048828, | |
| "learning_rate": 3.7313432835820893e-06, | |
| "loss": 1.9649, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.35714285714285715, | |
| "eval_loss": 1.7548341751098633, | |
| "eval_runtime": 12.6786, | |
| "eval_samples_per_second": 5.521, | |
| "eval_steps_per_second": 1.42, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 75, | |
| "total_flos": 3.73846047031296e+17, | |
| "train_loss": 2.0365064112345377, | |
| "train_runtime": 139.8497, | |
| "train_samples_per_second": 2.145, | |
| "train_steps_per_second": 0.536 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.27741935483870966, | |
| "eval_loss": 1.9401741027832031, | |
| "eval_runtime": 29.4337, | |
| "eval_samples_per_second": 5.266, | |
| "eval_steps_per_second": 1.325, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.27741935483870966, | |
| "eval_loss": 1.9401745796203613, | |
| "eval_runtime": 28.1588, | |
| "eval_samples_per_second": 5.505, | |
| "eval_steps_per_second": 1.385, | |
| "step": 75 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 75, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 3.73846047031296e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |