| { |
| "best_metric": 0.3903774321079254, |
| "best_model_checkpoint": "./vit-base-augmented-ph2-patch-32/checkpoint-50", |
| "epoch": 4.0, |
| "global_step": 552, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.07, |
| "learning_rate": 0.00019637681159420293, |
| "loss": 0.4337, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.14, |
| "learning_rate": 0.0001927536231884058, |
| "loss": 0.1958, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.22, |
| "learning_rate": 0.0001891304347826087, |
| "loss": 0.2358, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.29, |
| "learning_rate": 0.0001855072463768116, |
| "loss": 0.1231, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.36, |
| "learning_rate": 0.0001818840579710145, |
| "loss": 0.1087, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.36, |
| "eval_accuracy": 0.8683760683760684, |
| "eval_f1": 0.8683760683760683, |
| "eval_loss": 0.3903774321079254, |
| "eval_precision": 0.8683760683760684, |
| "eval_recall": 0.8683760683760684, |
| "eval_runtime": 10.4647, |
| "eval_samples_per_second": 55.902, |
| "eval_steps_per_second": 7.071, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.43, |
| "learning_rate": 0.0001782608695652174, |
| "loss": 0.0318, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.51, |
| "learning_rate": 0.0001746376811594203, |
| "loss": 0.0115, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.58, |
| "learning_rate": 0.0001710144927536232, |
| "loss": 0.0412, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.65, |
| "learning_rate": 0.0001673913043478261, |
| "loss": 0.009, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.72, |
| "learning_rate": 0.000163768115942029, |
| "loss": 0.066, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.72, |
| "eval_accuracy": 0.8273504273504273, |
| "eval_f1": 0.8273504273504273, |
| "eval_loss": 0.7072790265083313, |
| "eval_precision": 0.8273504273504273, |
| "eval_recall": 0.8273504273504273, |
| "eval_runtime": 10.5596, |
| "eval_samples_per_second": 55.4, |
| "eval_steps_per_second": 7.008, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.8, |
| "learning_rate": 0.00016014492753623188, |
| "loss": 0.0898, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.87, |
| "learning_rate": 0.0001565217391304348, |
| "loss": 0.0228, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.94, |
| "learning_rate": 0.00015289855072463769, |
| "loss": 0.035, |
| "step": 130 |
| }, |
| { |
| "epoch": 1.01, |
| "learning_rate": 0.00014927536231884058, |
| "loss": 0.0211, |
| "step": 140 |
| }, |
| { |
| "epoch": 1.09, |
| "learning_rate": 0.0001456521739130435, |
| "loss": 0.0092, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.09, |
| "eval_accuracy": 0.8153846153846154, |
| "eval_f1": 0.8153846153846154, |
| "eval_loss": 0.663502037525177, |
| "eval_precision": 0.8153846153846154, |
| "eval_recall": 0.8153846153846154, |
| "eval_runtime": 10.5172, |
| "eval_samples_per_second": 55.623, |
| "eval_steps_per_second": 7.036, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.16, |
| "learning_rate": 0.00014202898550724638, |
| "loss": 0.006, |
| "step": 160 |
| }, |
| { |
| "epoch": 1.23, |
| "learning_rate": 0.00013840579710144927, |
| "loss": 0.0686, |
| "step": 170 |
| }, |
| { |
| "epoch": 1.3, |
| "learning_rate": 0.0001347826086956522, |
| "loss": 0.004, |
| "step": 180 |
| }, |
| { |
| "epoch": 1.38, |
| "learning_rate": 0.00013115942028985508, |
| "loss": 0.0654, |
| "step": 190 |
| }, |
| { |
| "epoch": 1.45, |
| "learning_rate": 0.00012753623188405797, |
| "loss": 0.0716, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.45, |
| "eval_accuracy": 0.8341880341880342, |
| "eval_f1": 0.8341880341880342, |
| "eval_loss": 0.7824386954307556, |
| "eval_precision": 0.8341880341880342, |
| "eval_recall": 0.8341880341880342, |
| "eval_runtime": 10.4823, |
| "eval_samples_per_second": 55.808, |
| "eval_steps_per_second": 7.06, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.52, |
| "learning_rate": 0.00012391304347826086, |
| "loss": 0.1392, |
| "step": 210 |
| }, |
| { |
| "epoch": 1.59, |
| "learning_rate": 0.00012028985507246378, |
| "loss": 0.0547, |
| "step": 220 |
| }, |
| { |
| "epoch": 1.67, |
| "learning_rate": 0.00011666666666666668, |
| "loss": 0.0113, |
| "step": 230 |
| }, |
| { |
| "epoch": 1.74, |
| "learning_rate": 0.00011304347826086956, |
| "loss": 0.004, |
| "step": 240 |
| }, |
| { |
| "epoch": 1.81, |
| "learning_rate": 0.00010942028985507247, |
| "loss": 0.0056, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.81, |
| "eval_accuracy": 0.8957264957264958, |
| "eval_f1": 0.8957264957264957, |
| "eval_loss": 0.5070696473121643, |
| "eval_precision": 0.8957264957264958, |
| "eval_recall": 0.8957264957264958, |
| "eval_runtime": 10.5158, |
| "eval_samples_per_second": 55.63, |
| "eval_steps_per_second": 7.037, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.88, |
| "learning_rate": 0.00010579710144927538, |
| "loss": 0.0037, |
| "step": 260 |
| }, |
| { |
| "epoch": 1.96, |
| "learning_rate": 0.00010217391304347828, |
| "loss": 0.0027, |
| "step": 270 |
| }, |
| { |
| "epoch": 2.03, |
| "learning_rate": 9.855072463768117e-05, |
| "loss": 0.0025, |
| "step": 280 |
| }, |
| { |
| "epoch": 2.1, |
| "learning_rate": 9.492753623188406e-05, |
| "loss": 0.0024, |
| "step": 290 |
| }, |
| { |
| "epoch": 2.17, |
| "learning_rate": 9.130434782608696e-05, |
| "loss": 0.0023, |
| "step": 300 |
| }, |
| { |
| "epoch": 2.17, |
| "eval_accuracy": 0.8854700854700854, |
| "eval_f1": 0.8854700854700854, |
| "eval_loss": 0.5977781414985657, |
| "eval_precision": 0.8854700854700854, |
| "eval_recall": 0.8854700854700854, |
| "eval_runtime": 10.6417, |
| "eval_samples_per_second": 54.973, |
| "eval_steps_per_second": 6.954, |
| "step": 300 |
| }, |
| { |
| "epoch": 2.25, |
| "learning_rate": 8.768115942028987e-05, |
| "loss": 0.0022, |
| "step": 310 |
| }, |
| { |
| "epoch": 2.32, |
| "learning_rate": 8.405797101449276e-05, |
| "loss": 0.0021, |
| "step": 320 |
| }, |
| { |
| "epoch": 2.39, |
| "learning_rate": 8.043478260869566e-05, |
| "loss": 0.002, |
| "step": 330 |
| }, |
| { |
| "epoch": 2.46, |
| "learning_rate": 7.681159420289855e-05, |
| "loss": 0.0019, |
| "step": 340 |
| }, |
| { |
| "epoch": 2.54, |
| "learning_rate": 7.318840579710145e-05, |
| "loss": 0.0019, |
| "step": 350 |
| }, |
| { |
| "epoch": 2.54, |
| "eval_accuracy": 0.8854700854700854, |
| "eval_f1": 0.8854700854700854, |
| "eval_loss": 0.614289402961731, |
| "eval_precision": 0.8854700854700854, |
| "eval_recall": 0.8854700854700854, |
| "eval_runtime": 10.5592, |
| "eval_samples_per_second": 55.402, |
| "eval_steps_per_second": 7.008, |
| "step": 350 |
| }, |
| { |
| "epoch": 2.61, |
| "learning_rate": 6.956521739130436e-05, |
| "loss": 0.0018, |
| "step": 360 |
| }, |
| { |
| "epoch": 2.68, |
| "learning_rate": 6.594202898550726e-05, |
| "loss": 0.0018, |
| "step": 370 |
| }, |
| { |
| "epoch": 2.75, |
| "learning_rate": 6.231884057971015e-05, |
| "loss": 0.0017, |
| "step": 380 |
| }, |
| { |
| "epoch": 2.83, |
| "learning_rate": 5.869565217391305e-05, |
| "loss": 0.0017, |
| "step": 390 |
| }, |
| { |
| "epoch": 2.9, |
| "learning_rate": 5.507246376811594e-05, |
| "loss": 0.0016, |
| "step": 400 |
| }, |
| { |
| "epoch": 2.9, |
| "eval_accuracy": 0.8888888888888888, |
| "eval_f1": 0.8888888888888888, |
| "eval_loss": 0.6227333545684814, |
| "eval_precision": 0.8888888888888888, |
| "eval_recall": 0.8888888888888888, |
| "eval_runtime": 10.6134, |
| "eval_samples_per_second": 55.119, |
| "eval_steps_per_second": 6.972, |
| "step": 400 |
| }, |
| { |
| "epoch": 2.97, |
| "learning_rate": 5.144927536231884e-05, |
| "loss": 0.0016, |
| "step": 410 |
| }, |
| { |
| "epoch": 3.04, |
| "learning_rate": 4.782608695652174e-05, |
| "loss": 0.0016, |
| "step": 420 |
| }, |
| { |
| "epoch": 3.12, |
| "learning_rate": 4.4202898550724645e-05, |
| "loss": 0.0015, |
| "step": 430 |
| }, |
| { |
| "epoch": 3.19, |
| "learning_rate": 4.057971014492754e-05, |
| "loss": 0.0016, |
| "step": 440 |
| }, |
| { |
| "epoch": 3.26, |
| "learning_rate": 3.695652173913043e-05, |
| "loss": 0.0015, |
| "step": 450 |
| }, |
| { |
| "epoch": 3.26, |
| "eval_accuracy": 0.8888888888888888, |
| "eval_f1": 0.8888888888888888, |
| "eval_loss": 0.6293916702270508, |
| "eval_precision": 0.8888888888888888, |
| "eval_recall": 0.8888888888888888, |
| "eval_runtime": 10.4957, |
| "eval_samples_per_second": 55.737, |
| "eval_steps_per_second": 7.05, |
| "step": 450 |
| }, |
| { |
| "epoch": 3.33, |
| "learning_rate": 3.3333333333333335e-05, |
| "loss": 0.0015, |
| "step": 460 |
| }, |
| { |
| "epoch": 3.41, |
| "learning_rate": 2.971014492753623e-05, |
| "loss": 0.0015, |
| "step": 470 |
| }, |
| { |
| "epoch": 3.48, |
| "learning_rate": 2.608695652173913e-05, |
| "loss": 0.0015, |
| "step": 480 |
| }, |
| { |
| "epoch": 3.55, |
| "learning_rate": 2.246376811594203e-05, |
| "loss": 0.0014, |
| "step": 490 |
| }, |
| { |
| "epoch": 3.62, |
| "learning_rate": 1.8840579710144928e-05, |
| "loss": 0.0014, |
| "step": 500 |
| }, |
| { |
| "epoch": 3.62, |
| "eval_accuracy": 0.8888888888888888, |
| "eval_f1": 0.8888888888888888, |
| "eval_loss": 0.6337538957595825, |
| "eval_precision": 0.8888888888888888, |
| "eval_recall": 0.8888888888888888, |
| "eval_runtime": 10.5742, |
| "eval_samples_per_second": 55.323, |
| "eval_steps_per_second": 6.998, |
| "step": 500 |
| }, |
| { |
| "epoch": 3.7, |
| "learning_rate": 1.5217391304347828e-05, |
| "loss": 0.0014, |
| "step": 510 |
| }, |
| { |
| "epoch": 3.77, |
| "learning_rate": 1.1594202898550725e-05, |
| "loss": 0.0014, |
| "step": 520 |
| }, |
| { |
| "epoch": 3.84, |
| "learning_rate": 7.971014492753623e-06, |
| "loss": 0.0014, |
| "step": 530 |
| }, |
| { |
| "epoch": 3.91, |
| "learning_rate": 4.347826086956522e-06, |
| "loss": 0.0014, |
| "step": 540 |
| }, |
| { |
| "epoch": 3.99, |
| "learning_rate": 7.246376811594203e-07, |
| "loss": 0.0014, |
| "step": 550 |
| }, |
| { |
| "epoch": 3.99, |
| "eval_accuracy": 0.8888888888888888, |
| "eval_f1": 0.8888888888888888, |
| "eval_loss": 0.6350913643836975, |
| "eval_precision": 0.8888888888888888, |
| "eval_recall": 0.8888888888888888, |
| "eval_runtime": 10.4521, |
| "eval_samples_per_second": 55.97, |
| "eval_steps_per_second": 7.08, |
| "step": 550 |
| }, |
| { |
| "epoch": 4.0, |
| "step": 552, |
| "total_flos": 6.95095773216768e+17, |
| "train_loss": 0.034777708027673805, |
| "train_runtime": 468.7352, |
| "train_samples_per_second": 18.774, |
| "train_steps_per_second": 1.178 |
| } |
| ], |
| "max_steps": 552, |
| "num_train_epochs": 4, |
| "total_flos": 6.95095773216768e+17, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
