| { | |
| "best_metric": 0.8994708994708994, | |
| "best_model_checkpoint": "Har_Finetuned-ViT-Hybrid/checkpoint-1002", | |
| "epoch": 5.997014925373135, | |
| "global_step": 1002, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4.950495049504951e-06, | |
| "loss": 2.8031, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.900990099009901e-06, | |
| "loss": 2.5937, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1.4851485148514851e-05, | |
| "loss": 2.2631, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.9801980198019803e-05, | |
| "loss": 1.8858, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 2.4752475247524754e-05, | |
| "loss": 1.5579, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 2.9702970297029702e-05, | |
| "loss": 1.1878, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 3.465346534653465e-05, | |
| "loss": 0.9966, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 3.9603960396039605e-05, | |
| "loss": 0.9343, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 4.455445544554456e-05, | |
| "loss": 0.897, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.950495049504951e-05, | |
| "loss": 0.8911, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.9500554938956716e-05, | |
| "loss": 0.8231, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 4.894561598224195e-05, | |
| "loss": 0.8594, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.8390677025527196e-05, | |
| "loss": 0.7903, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 4.783573806881244e-05, | |
| "loss": 0.7901, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 4.7280799112097675e-05, | |
| "loss": 0.7695, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 4.672586015538291e-05, | |
| "loss": 0.7923, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8698412698412699, | |
| "eval_loss": 0.44196292757987976, | |
| "eval_runtime": 70.4064, | |
| "eval_samples_per_second": 26.844, | |
| "eval_steps_per_second": 1.69, | |
| "step": 167 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 4.617092119866815e-05, | |
| "loss": 0.6929, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 4.5615982241953385e-05, | |
| "loss": 0.6409, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 4.506104328523862e-05, | |
| "loss": 0.6611, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 4.4506104328523865e-05, | |
| "loss": 0.6694, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 4.395116537180911e-05, | |
| "loss": 0.622, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 4.3396226415094345e-05, | |
| "loss": 0.6312, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 4.284128745837958e-05, | |
| "loss": 0.6552, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 4.228634850166482e-05, | |
| "loss": 0.6335, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 4.1731409544950054e-05, | |
| "loss": 0.6952, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 4.11764705882353e-05, | |
| "loss": 0.5943, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 4.062153163152054e-05, | |
| "loss": 0.5843, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 4.006659267480578e-05, | |
| "loss": 0.6073, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 3.9511653718091014e-05, | |
| "loss": 0.6046, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 3.895671476137625e-05, | |
| "loss": 0.5643, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 3.8401775804661487e-05, | |
| "loss": 0.6454, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 3.784683684794672e-05, | |
| "loss": 0.5922, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 3.7291897891231966e-05, | |
| "loss": 0.5555, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.882010582010582, | |
| "eval_loss": 0.3810536563396454, | |
| "eval_runtime": 70.4153, | |
| "eval_samples_per_second": 26.841, | |
| "eval_steps_per_second": 1.69, | |
| "step": 334 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 3.673695893451721e-05, | |
| "loss": 0.6331, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 3.6182019977802446e-05, | |
| "loss": 0.4524, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 3.562708102108768e-05, | |
| "loss": 0.4958, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 3.507214206437292e-05, | |
| "loss": 0.5111, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 3.4517203107658156e-05, | |
| "loss": 0.5885, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 3.39622641509434e-05, | |
| "loss": 0.5497, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 3.340732519422864e-05, | |
| "loss": 0.4644, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 3.285238623751388e-05, | |
| "loss": 0.5118, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 3.2297447280799115e-05, | |
| "loss": 0.5281, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 3.174250832408435e-05, | |
| "loss": 0.5025, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 3.118756936736959e-05, | |
| "loss": 0.4469, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 3.0632630410654825e-05, | |
| "loss": 0.4594, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 3.0077691453940065e-05, | |
| "loss": 0.5019, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 2.9522752497225308e-05, | |
| "loss": 0.5002, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 2.8967813540510548e-05, | |
| "loss": 0.4876, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 2.8412874583795784e-05, | |
| "loss": 0.5479, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 2.785793562708102e-05, | |
| "loss": 0.4734, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8957671957671958, | |
| "eval_loss": 0.34475043416023254, | |
| "eval_runtime": 70.2044, | |
| "eval_samples_per_second": 26.921, | |
| "eval_steps_per_second": 1.695, | |
| "step": 501 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 2.730299667036626e-05, | |
| "loss": 0.4618, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 2.6748057713651497e-05, | |
| "loss": 0.4161, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 2.619311875693674e-05, | |
| "loss": 0.3976, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 2.563817980022198e-05, | |
| "loss": 0.3763, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 2.5083240843507217e-05, | |
| "loss": 0.4015, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 2.4528301886792453e-05, | |
| "loss": 0.3991, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 2.3973362930077693e-05, | |
| "loss": 0.4204, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 2.3418423973362933e-05, | |
| "loss": 0.3972, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 2.286348501664817e-05, | |
| "loss": 0.3807, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 2.230854605993341e-05, | |
| "loss": 0.4301, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 2.175360710321865e-05, | |
| "loss": 0.4949, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 2.1198668146503886e-05, | |
| "loss": 0.3774, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 2.0643729189789122e-05, | |
| "loss": 0.3323, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 2.0088790233074362e-05, | |
| "loss": 0.4668, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 1.9533851276359602e-05, | |
| "loss": 0.4075, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 1.897891231964484e-05, | |
| "loss": 0.4019, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8925925925925926, | |
| "eval_loss": 0.35210713744163513, | |
| "eval_runtime": 70.2455, | |
| "eval_samples_per_second": 26.906, | |
| "eval_steps_per_second": 1.694, | |
| "step": 668 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 1.842397336293008e-05, | |
| "loss": 0.4296, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 1.786903440621532e-05, | |
| "loss": 0.4051, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 1.7314095449500555e-05, | |
| "loss": 0.3749, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 1.6759156492785795e-05, | |
| "loss": 0.3502, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 1.6204217536071035e-05, | |
| "loss": 0.308, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 1.564927857935627e-05, | |
| "loss": 0.3514, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 1.509433962264151e-05, | |
| "loss": 0.3379, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "learning_rate": 1.4539400665926751e-05, | |
| "loss": 0.3101, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 1.3984461709211987e-05, | |
| "loss": 0.3375, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 1.3429522752497226e-05, | |
| "loss": 0.3195, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "learning_rate": 1.2874583795782464e-05, | |
| "loss": 0.3878, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 1.2319644839067704e-05, | |
| "loss": 0.3555, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 1.1764705882352942e-05, | |
| "loss": 0.3177, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 1.120976692563818e-05, | |
| "loss": 0.294, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "learning_rate": 1.0654827968923418e-05, | |
| "loss": 0.3676, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "learning_rate": 1.0099889012208658e-05, | |
| "loss": 0.3074, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 9.544950055493896e-06, | |
| "loss": 0.3622, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8925925925925926, | |
| "eval_loss": 0.350492000579834, | |
| "eval_runtime": 70.3191, | |
| "eval_samples_per_second": 26.877, | |
| "eval_steps_per_second": 1.692, | |
| "step": 835 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 8.990011098779134e-06, | |
| "loss": 0.3231, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "learning_rate": 8.435072142064374e-06, | |
| "loss": 0.3339, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.15, | |
| "learning_rate": 7.880133185349611e-06, | |
| "loss": 0.2668, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "learning_rate": 7.325194228634851e-06, | |
| "loss": 0.2918, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "learning_rate": 6.77025527192009e-06, | |
| "loss": 0.2817, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 6.215316315205328e-06, | |
| "loss": 0.269, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 5.39, | |
| "learning_rate": 5.660377358490566e-06, | |
| "loss": 0.3038, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 5.105438401775805e-06, | |
| "loss": 0.2663, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "learning_rate": 4.5504994450610434e-06, | |
| "loss": 0.2752, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "learning_rate": 3.9955604883462825e-06, | |
| "loss": 0.3432, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 5.63, | |
| "learning_rate": 3.4406215316315207e-06, | |
| "loss": 0.267, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 2.8856825749167593e-06, | |
| "loss": 0.3169, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "learning_rate": 2.330743618201998e-06, | |
| "loss": 0.2609, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 1.7758046614872365e-06, | |
| "loss": 0.3151, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 5.87, | |
| "learning_rate": 1.2208657047724751e-06, | |
| "loss": 0.2772, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 6.659267480577136e-07, | |
| "loss": 0.358, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "learning_rate": 1.1098779134295228e-07, | |
| "loss": 0.2921, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8994708994708994, | |
| "eval_loss": 0.3382543921470642, | |
| "eval_runtime": 70.2009, | |
| "eval_samples_per_second": 26.923, | |
| "eval_steps_per_second": 1.695, | |
| "step": 1002 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 1002, | |
| "total_flos": 1.6742089052991848e+19, | |
| "train_loss": 0.5761546168379679, | |
| "train_runtime": 6609.6541, | |
| "train_samples_per_second": 9.722, | |
| "train_steps_per_second": 0.152 | |
| } | |
| ], | |
| "max_steps": 1002, | |
| "num_train_epochs": 6, | |
| "total_flos": 1.6742089052991848e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |