{ "best_metric": 0.8984126984126984, "best_model_checkpoint": "Har_Finetuned-ViT-Hybrid_/checkpoint-835", "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": 2.8865, "step": 10 }, { "epoch": 0.12, "learning_rate": 9.900990099009901e-06, "loss": 2.7196, "step": 20 }, { "epoch": 0.18, "learning_rate": 1.4851485148514851e-05, "loss": 2.4283, "step": 30 }, { "epoch": 0.24, "learning_rate": 1.9801980198019803e-05, "loss": 2.0403, "step": 40 }, { "epoch": 0.3, "learning_rate": 2.4752475247524754e-05, "loss": 1.6041, "step": 50 }, { "epoch": 0.36, "learning_rate": 2.9702970297029702e-05, "loss": 1.2181, "step": 60 }, { "epoch": 0.42, "learning_rate": 3.465346534653465e-05, "loss": 0.991, "step": 70 }, { "epoch": 0.48, "learning_rate": 3.9603960396039605e-05, "loss": 0.9408, "step": 80 }, { "epoch": 0.54, "learning_rate": 4.455445544554456e-05, "loss": 0.925, "step": 90 }, { "epoch": 0.6, "learning_rate": 4.950495049504951e-05, "loss": 0.8281, "step": 100 }, { "epoch": 0.66, "learning_rate": 4.9500554938956716e-05, "loss": 0.7048, "step": 110 }, { "epoch": 0.72, "learning_rate": 4.894561598224195e-05, "loss": 0.7256, "step": 120 }, { "epoch": 0.78, "learning_rate": 4.8390677025527196e-05, "loss": 0.8498, "step": 130 }, { "epoch": 0.84, "learning_rate": 4.783573806881244e-05, "loss": 0.8356, "step": 140 }, { "epoch": 0.9, "learning_rate": 4.7280799112097675e-05, "loss": 0.827, "step": 150 }, { "epoch": 0.96, "learning_rate": 4.672586015538291e-05, "loss": 0.7647, "step": 160 }, { "epoch": 1.0, "eval_accuracy": 0.855026455026455, "eval_loss": 0.4917130470275879, "eval_runtime": 66.0948, "eval_samples_per_second": 28.595, "eval_steps_per_second": 1.8, "step": 167 }, { "epoch": 1.02, "learning_rate": 4.617092119866815e-05, "loss": 0.7168, "step": 170 }, { "epoch": 1.08, "learning_rate": 4.5615982241953385e-05, "loss": 0.6185, "step": 180 }, { "epoch": 1.14, "learning_rate": 4.506104328523862e-05, "loss": 0.6362, "step": 190 }, { "epoch": 1.2, "learning_rate": 4.4506104328523865e-05, "loss": 0.6327, "step": 200 }, { "epoch": 1.26, "learning_rate": 4.395116537180911e-05, "loss": 0.6509, "step": 210 }, { "epoch": 1.32, "learning_rate": 4.3396226415094345e-05, "loss": 0.6149, "step": 220 }, { "epoch": 1.38, "learning_rate": 4.284128745837958e-05, "loss": 0.6323, "step": 230 }, { "epoch": 1.44, "learning_rate": 4.228634850166482e-05, "loss": 0.5855, "step": 240 }, { "epoch": 1.5, "learning_rate": 4.1731409544950054e-05, "loss": 0.6605, "step": 250 }, { "epoch": 1.56, "learning_rate": 4.11764705882353e-05, "loss": 0.597, "step": 260 }, { "epoch": 1.61, "learning_rate": 4.062153163152054e-05, "loss": 0.6602, "step": 270 }, { "epoch": 1.67, "learning_rate": 4.006659267480578e-05, "loss": 0.6923, "step": 280 }, { "epoch": 1.73, "learning_rate": 3.9511653718091014e-05, "loss": 0.6574, "step": 290 }, { "epoch": 1.79, "learning_rate": 3.895671476137625e-05, "loss": 0.5841, "step": 300 }, { "epoch": 1.85, "learning_rate": 3.8401775804661487e-05, "loss": 0.5439, "step": 310 }, { "epoch": 1.91, "learning_rate": 3.784683684794672e-05, "loss": 0.508, "step": 320 }, { "epoch": 1.97, "learning_rate": 3.7291897891231966e-05, "loss": 0.6376, "step": 330 }, { "epoch": 2.0, "eval_accuracy": 0.882010582010582, "eval_loss": 0.36341825127601624, "eval_runtime": 65.6235, "eval_samples_per_second": 28.801, "eval_steps_per_second": 1.813, "step": 334 }, { "epoch": 2.04, "learning_rate": 3.673695893451721e-05, "loss": 0.6268, "step": 340 }, { "epoch": 2.1, "learning_rate": 3.6182019977802446e-05, "loss": 0.5205, "step": 350 }, { "epoch": 2.16, "learning_rate": 3.562708102108768e-05, "loss": 0.5141, "step": 360 }, { "epoch": 2.21, "learning_rate": 3.507214206437292e-05, "loss": 0.4614, "step": 370 }, { "epoch": 2.27, "learning_rate": 3.4517203107658156e-05, "loss": 0.56, "step": 380 }, { "epoch": 2.33, "learning_rate": 3.39622641509434e-05, "loss": 0.5434, "step": 390 }, { "epoch": 2.39, "learning_rate": 3.340732519422864e-05, "loss": 0.5106, "step": 400 }, { "epoch": 2.45, "learning_rate": 3.285238623751388e-05, "loss": 0.4774, "step": 410 }, { "epoch": 2.51, "learning_rate": 3.2297447280799115e-05, "loss": 0.4966, "step": 420 }, { "epoch": 2.57, "learning_rate": 3.174250832408435e-05, "loss": 0.4602, "step": 430 }, { "epoch": 2.63, "learning_rate": 3.118756936736959e-05, "loss": 0.5058, "step": 440 }, { "epoch": 2.69, "learning_rate": 3.0632630410654825e-05, "loss": 0.5123, "step": 450 }, { "epoch": 2.75, "learning_rate": 3.0077691453940065e-05, "loss": 0.5248, "step": 460 }, { "epoch": 2.81, "learning_rate": 2.9522752497225308e-05, "loss": 0.4786, "step": 470 }, { "epoch": 2.87, "learning_rate": 2.8967813540510548e-05, "loss": 0.5099, "step": 480 }, { "epoch": 2.93, "learning_rate": 2.8412874583795784e-05, "loss": 0.5054, "step": 490 }, { "epoch": 2.99, "learning_rate": 2.785793562708102e-05, "loss": 0.5051, "step": 500 }, { "epoch": 3.0, "eval_accuracy": 0.8915343915343915, "eval_loss": 0.33456629514694214, "eval_runtime": 65.9125, "eval_samples_per_second": 28.674, "eval_steps_per_second": 1.805, "step": 501 }, { "epoch": 3.05, "learning_rate": 2.730299667036626e-05, "loss": 0.4555, "step": 510 }, { "epoch": 3.11, "learning_rate": 2.6748057713651497e-05, "loss": 0.4846, "step": 520 }, { "epoch": 3.17, "learning_rate": 2.619311875693674e-05, "loss": 0.3832, "step": 530 }, { "epoch": 3.23, "learning_rate": 2.563817980022198e-05, "loss": 0.4917, "step": 540 }, { "epoch": 3.29, "learning_rate": 2.5083240843507217e-05, "loss": 0.3939, "step": 550 }, { "epoch": 3.35, "learning_rate": 2.4528301886792453e-05, "loss": 0.4226, "step": 560 }, { "epoch": 3.41, "learning_rate": 2.3973362930077693e-05, "loss": 0.4191, "step": 570 }, { "epoch": 3.47, "learning_rate": 2.3418423973362933e-05, "loss": 0.4151, "step": 580 }, { "epoch": 3.53, "learning_rate": 2.286348501664817e-05, "loss": 0.3625, "step": 590 }, { "epoch": 3.59, "learning_rate": 2.230854605993341e-05, "loss": 0.438, "step": 600 }, { "epoch": 3.65, "learning_rate": 2.175360710321865e-05, "loss": 0.3733, "step": 610 }, { "epoch": 3.71, "learning_rate": 2.1198668146503886e-05, "loss": 0.4121, "step": 620 }, { "epoch": 3.77, "learning_rate": 2.0643729189789122e-05, "loss": 0.4718, "step": 630 }, { "epoch": 3.83, "learning_rate": 2.0088790233074362e-05, "loss": 0.4434, "step": 640 }, { "epoch": 3.89, "learning_rate": 1.9533851276359602e-05, "loss": 0.4075, "step": 650 }, { "epoch": 3.95, "learning_rate": 1.897891231964484e-05, "loss": 0.3472, "step": 660 }, { "epoch": 4.0, "eval_accuracy": 0.8978835978835978, "eval_loss": 0.32800114154815674, "eval_runtime": 65.7835, "eval_samples_per_second": 28.731, "eval_steps_per_second": 1.809, "step": 668 }, { "epoch": 4.01, "learning_rate": 1.842397336293008e-05, "loss": 0.421, "step": 670 }, { "epoch": 4.07, "learning_rate": 1.786903440621532e-05, "loss": 0.3567, "step": 680 }, { "epoch": 4.13, "learning_rate": 1.7314095449500555e-05, "loss": 0.3756, "step": 690 }, { "epoch": 4.19, "learning_rate": 1.6759156492785795e-05, "loss": 0.4052, "step": 700 }, { "epoch": 4.25, "learning_rate": 1.6204217536071035e-05, "loss": 0.3426, "step": 710 }, { "epoch": 4.31, "learning_rate": 1.564927857935627e-05, "loss": 0.3318, "step": 720 }, { "epoch": 4.37, "learning_rate": 1.509433962264151e-05, "loss": 0.3351, "step": 730 }, { "epoch": 4.43, "learning_rate": 1.4539400665926751e-05, "loss": 0.3338, "step": 740 }, { "epoch": 4.49, "learning_rate": 1.3984461709211987e-05, "loss": 0.3599, "step": 750 }, { "epoch": 4.55, "learning_rate": 1.3429522752497226e-05, "loss": 0.3168, "step": 760 }, { "epoch": 4.61, "learning_rate": 1.2874583795782464e-05, "loss": 0.3367, "step": 770 }, { "epoch": 4.67, "learning_rate": 1.2319644839067704e-05, "loss": 0.3331, "step": 780 }, { "epoch": 4.73, "learning_rate": 1.1764705882352942e-05, "loss": 0.3585, "step": 790 }, { "epoch": 4.79, "learning_rate": 1.120976692563818e-05, "loss": 0.3229, "step": 800 }, { "epoch": 4.85, "learning_rate": 1.0654827968923418e-05, "loss": 0.365, "step": 810 }, { "epoch": 4.91, "learning_rate": 1.0099889012208658e-05, "loss": 0.3534, "step": 820 }, { "epoch": 4.97, "learning_rate": 9.544950055493896e-06, "loss": 0.2982, "step": 830 }, { "epoch": 5.0, "eval_accuracy": 0.8984126984126984, "eval_loss": 0.32516443729400635, "eval_runtime": 65.6979, "eval_samples_per_second": 28.768, "eval_steps_per_second": 1.811, "step": 835 }, { "epoch": 5.03, "learning_rate": 8.990011098779134e-06, "loss": 0.3674, "step": 840 }, { "epoch": 5.09, "learning_rate": 8.435072142064374e-06, "loss": 0.325, "step": 850 }, { "epoch": 5.15, "learning_rate": 7.880133185349611e-06, "loss": 0.3409, "step": 860 }, { "epoch": 5.21, "learning_rate": 7.325194228634851e-06, "loss": 0.359, "step": 870 }, { "epoch": 5.27, "learning_rate": 6.77025527192009e-06, "loss": 0.2572, "step": 880 }, { "epoch": 5.33, "learning_rate": 6.215316315205328e-06, "loss": 0.2866, "step": 890 }, { "epoch": 5.39, "learning_rate": 5.660377358490566e-06, "loss": 0.3735, "step": 900 }, { "epoch": 5.45, "learning_rate": 5.105438401775805e-06, "loss": 0.3473, "step": 910 }, { "epoch": 5.51, "learning_rate": 4.5504994450610434e-06, "loss": 0.336, "step": 920 }, { "epoch": 5.57, "learning_rate": 3.9955604883462825e-06, "loss": 0.29, "step": 930 }, { "epoch": 5.63, "learning_rate": 3.4406215316315207e-06, "loss": 0.3305, "step": 940 }, { "epoch": 5.69, "learning_rate": 2.8856825749167593e-06, "loss": 0.2974, "step": 950 }, { "epoch": 5.75, "learning_rate": 2.330743618201998e-06, "loss": 0.2603, "step": 960 }, { "epoch": 5.81, "learning_rate": 1.7758046614872365e-06, "loss": 0.2109, "step": 970 }, { "epoch": 5.87, "learning_rate": 1.2208657047724751e-06, "loss": 0.2934, "step": 980 }, { "epoch": 5.93, "learning_rate": 6.659267480577136e-07, "loss": 0.2565, "step": 990 }, { "epoch": 5.99, "learning_rate": 1.1098779134295228e-07, "loss": 0.319, "step": 1000 }, { "epoch": 6.0, "eval_accuracy": 0.8978835978835978, "eval_loss": 0.32160982489585876, "eval_runtime": 65.7629, "eval_samples_per_second": 28.74, "eval_steps_per_second": 1.81, "step": 1002 }, { "epoch": 6.0, "step": 1002, "total_flos": 1.6742089052991848e+19, "train_loss": 0.585026529556263, "train_runtime": 6271.0965, "train_samples_per_second": 10.247, "train_steps_per_second": 0.16 } ], "max_steps": 1002, "num_train_epochs": 6, "total_flos": 1.6742089052991848e+19, "trial_name": null, "trial_params": null }