| { |
| "best_metric": 0.6923076923076923, |
| "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-humeda-DAV7/checkpoint-180", |
| "epoch": 30.0, |
| "eval_steps": 500, |
| "global_step": 360, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.8333333333333334, |
| "grad_norm": 4.077209949493408, |
| "learning_rate": 5.555555555555557e-06, |
| "loss": 1.6286, |
| "step": 10 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.2692307692307692, |
| "eval_loss": 1.5624008178710938, |
| "eval_runtime": 0.9603, |
| "eval_samples_per_second": 54.152, |
| "eval_steps_per_second": 1.041, |
| "step": 12 |
| }, |
| { |
| "epoch": 1.6666666666666665, |
| "grad_norm": 3.5481019020080566, |
| "learning_rate": 1.1111111111111113e-05, |
| "loss": 1.5581, |
| "step": 20 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.5192307692307693, |
| "eval_loss": 1.4705888032913208, |
| "eval_runtime": 0.9518, |
| "eval_samples_per_second": 54.635, |
| "eval_steps_per_second": 1.051, |
| "step": 24 |
| }, |
| { |
| "epoch": 2.5, |
| "grad_norm": 4.507537364959717, |
| "learning_rate": 1.6666666666666667e-05, |
| "loss": 1.4708, |
| "step": 30 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.5384615384615384, |
| "eval_loss": 1.2967134714126587, |
| "eval_runtime": 0.9985, |
| "eval_samples_per_second": 52.079, |
| "eval_steps_per_second": 1.002, |
| "step": 36 |
| }, |
| { |
| "epoch": 3.3333333333333335, |
| "grad_norm": 5.175262928009033, |
| "learning_rate": 1.9753086419753087e-05, |
| "loss": 1.3022, |
| "step": 40 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.46153846153846156, |
| "eval_loss": 1.1556154489517212, |
| "eval_runtime": 1.079, |
| "eval_samples_per_second": 48.194, |
| "eval_steps_per_second": 0.927, |
| "step": 48 |
| }, |
| { |
| "epoch": 4.166666666666667, |
| "grad_norm": 8.343924522399902, |
| "learning_rate": 1.9135802469135804e-05, |
| "loss": 1.0542, |
| "step": 50 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 17.81778907775879, |
| "learning_rate": 1.851851851851852e-05, |
| "loss": 0.9735, |
| "step": 60 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.5384615384615384, |
| "eval_loss": 1.035403847694397, |
| "eval_runtime": 0.9995, |
| "eval_samples_per_second": 52.024, |
| "eval_steps_per_second": 1.0, |
| "step": 60 |
| }, |
| { |
| "epoch": 5.833333333333333, |
| "grad_norm": 9.504965782165527, |
| "learning_rate": 1.7901234567901236e-05, |
| "loss": 0.7914, |
| "step": 70 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.4423076923076923, |
| "eval_loss": 1.1014288663864136, |
| "eval_runtime": 1.2662, |
| "eval_samples_per_second": 41.067, |
| "eval_steps_per_second": 0.79, |
| "step": 72 |
| }, |
| { |
| "epoch": 6.666666666666667, |
| "grad_norm": 12.34647274017334, |
| "learning_rate": 1.728395061728395e-05, |
| "loss": 0.7376, |
| "step": 80 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.4807692307692308, |
| "eval_loss": 1.108736276626587, |
| "eval_runtime": 0.9872, |
| "eval_samples_per_second": 52.676, |
| "eval_steps_per_second": 1.013, |
| "step": 84 |
| }, |
| { |
| "epoch": 7.5, |
| "grad_norm": 7.576508522033691, |
| "learning_rate": 1.6666666666666667e-05, |
| "loss": 0.692, |
| "step": 90 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.6346153846153846, |
| "eval_loss": 0.9630454182624817, |
| "eval_runtime": 0.9885, |
| "eval_samples_per_second": 52.605, |
| "eval_steps_per_second": 1.012, |
| "step": 96 |
| }, |
| { |
| "epoch": 8.333333333333334, |
| "grad_norm": 16.95194435119629, |
| "learning_rate": 1.6049382716049385e-05, |
| "loss": 0.6586, |
| "step": 100 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.6346153846153846, |
| "eval_loss": 0.9428613185882568, |
| "eval_runtime": 1.1109, |
| "eval_samples_per_second": 46.811, |
| "eval_steps_per_second": 0.9, |
| "step": 108 |
| }, |
| { |
| "epoch": 9.166666666666666, |
| "grad_norm": 10.81656265258789, |
| "learning_rate": 1.54320987654321e-05, |
| "loss": 0.618, |
| "step": 110 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 15.426957130432129, |
| "learning_rate": 1.4814814814814815e-05, |
| "loss": 0.5799, |
| "step": 120 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_accuracy": 0.6346153846153846, |
| "eval_loss": 0.9332149028778076, |
| "eval_runtime": 1.014, |
| "eval_samples_per_second": 51.283, |
| "eval_steps_per_second": 0.986, |
| "step": 120 |
| }, |
| { |
| "epoch": 10.833333333333334, |
| "grad_norm": 16.087358474731445, |
| "learning_rate": 1.4197530864197532e-05, |
| "loss": 0.5557, |
| "step": 130 |
| }, |
| { |
| "epoch": 11.0, |
| "eval_accuracy": 0.5192307692307693, |
| "eval_loss": 1.17124342918396, |
| "eval_runtime": 1.1766, |
| "eval_samples_per_second": 44.197, |
| "eval_steps_per_second": 0.85, |
| "step": 132 |
| }, |
| { |
| "epoch": 11.666666666666666, |
| "grad_norm": 13.390655517578125, |
| "learning_rate": 1.3580246913580248e-05, |
| "loss": 0.5233, |
| "step": 140 |
| }, |
| { |
| "epoch": 12.0, |
| "eval_accuracy": 0.5576923076923077, |
| "eval_loss": 1.0446679592132568, |
| "eval_runtime": 0.9935, |
| "eval_samples_per_second": 52.34, |
| "eval_steps_per_second": 1.007, |
| "step": 144 |
| }, |
| { |
| "epoch": 12.5, |
| "grad_norm": 13.491952896118164, |
| "learning_rate": 1.2962962962962964e-05, |
| "loss": 0.4427, |
| "step": 150 |
| }, |
| { |
| "epoch": 13.0, |
| "eval_accuracy": 0.6538461538461539, |
| "eval_loss": 0.8927649855613708, |
| "eval_runtime": 0.977, |
| "eval_samples_per_second": 53.224, |
| "eval_steps_per_second": 1.024, |
| "step": 156 |
| }, |
| { |
| "epoch": 13.333333333333334, |
| "grad_norm": 11.844417572021484, |
| "learning_rate": 1.234567901234568e-05, |
| "loss": 0.5043, |
| "step": 160 |
| }, |
| { |
| "epoch": 14.0, |
| "eval_accuracy": 0.5961538461538461, |
| "eval_loss": 1.012035846710205, |
| "eval_runtime": 1.0234, |
| "eval_samples_per_second": 50.812, |
| "eval_steps_per_second": 0.977, |
| "step": 168 |
| }, |
| { |
| "epoch": 14.166666666666666, |
| "grad_norm": 9.26688289642334, |
| "learning_rate": 1.1728395061728398e-05, |
| "loss": 0.4425, |
| "step": 170 |
| }, |
| { |
| "epoch": 15.0, |
| "grad_norm": 22.588224411010742, |
| "learning_rate": 1.1111111111111113e-05, |
| "loss": 0.4167, |
| "step": 180 |
| }, |
| { |
| "epoch": 15.0, |
| "eval_accuracy": 0.6923076923076923, |
| "eval_loss": 0.924092173576355, |
| "eval_runtime": 0.9867, |
| "eval_samples_per_second": 52.701, |
| "eval_steps_per_second": 1.013, |
| "step": 180 |
| }, |
| { |
| "epoch": 15.833333333333334, |
| "grad_norm": 12.42056655883789, |
| "learning_rate": 1.0493827160493827e-05, |
| "loss": 0.4601, |
| "step": 190 |
| }, |
| { |
| "epoch": 16.0, |
| "eval_accuracy": 0.6538461538461539, |
| "eval_loss": 0.884823203086853, |
| "eval_runtime": 1.1645, |
| "eval_samples_per_second": 44.653, |
| "eval_steps_per_second": 0.859, |
| "step": 192 |
| }, |
| { |
| "epoch": 16.666666666666668, |
| "grad_norm": 19.14876937866211, |
| "learning_rate": 9.876543209876543e-06, |
| "loss": 0.4619, |
| "step": 200 |
| }, |
| { |
| "epoch": 17.0, |
| "eval_accuracy": 0.6923076923076923, |
| "eval_loss": 0.9238922595977783, |
| "eval_runtime": 0.9708, |
| "eval_samples_per_second": 53.563, |
| "eval_steps_per_second": 1.03, |
| "step": 204 |
| }, |
| { |
| "epoch": 17.5, |
| "grad_norm": 11.291189193725586, |
| "learning_rate": 9.25925925925926e-06, |
| "loss": 0.3822, |
| "step": 210 |
| }, |
| { |
| "epoch": 18.0, |
| "eval_accuracy": 0.6730769230769231, |
| "eval_loss": 0.9207645654678345, |
| "eval_runtime": 1.3734, |
| "eval_samples_per_second": 37.862, |
| "eval_steps_per_second": 0.728, |
| "step": 216 |
| }, |
| { |
| "epoch": 18.333333333333332, |
| "grad_norm": 13.084733963012695, |
| "learning_rate": 8.641975308641975e-06, |
| "loss": 0.3707, |
| "step": 220 |
| }, |
| { |
| "epoch": 19.0, |
| "eval_accuracy": 0.5769230769230769, |
| "eval_loss": 1.037381887435913, |
| "eval_runtime": 0.9939, |
| "eval_samples_per_second": 52.318, |
| "eval_steps_per_second": 1.006, |
| "step": 228 |
| }, |
| { |
| "epoch": 19.166666666666668, |
| "grad_norm": 13.402117729187012, |
| "learning_rate": 8.024691358024692e-06, |
| "loss": 0.3756, |
| "step": 230 |
| }, |
| { |
| "epoch": 20.0, |
| "grad_norm": 18.240755081176758, |
| "learning_rate": 7.4074074074074075e-06, |
| "loss": 0.365, |
| "step": 240 |
| }, |
| { |
| "epoch": 20.0, |
| "eval_accuracy": 0.6538461538461539, |
| "eval_loss": 0.9899800419807434, |
| "eval_runtime": 1.3813, |
| "eval_samples_per_second": 37.646, |
| "eval_steps_per_second": 0.724, |
| "step": 240 |
| }, |
| { |
| "epoch": 20.833333333333332, |
| "grad_norm": 10.653473854064941, |
| "learning_rate": 6.790123456790124e-06, |
| "loss": 0.3412, |
| "step": 250 |
| }, |
| { |
| "epoch": 21.0, |
| "eval_accuracy": 0.6538461538461539, |
| "eval_loss": 1.0541142225265503, |
| "eval_runtime": 0.9898, |
| "eval_samples_per_second": 52.537, |
| "eval_steps_per_second": 1.01, |
| "step": 252 |
| }, |
| { |
| "epoch": 21.666666666666668, |
| "grad_norm": 11.7351655960083, |
| "learning_rate": 6.17283950617284e-06, |
| "loss": 0.3265, |
| "step": 260 |
| }, |
| { |
| "epoch": 22.0, |
| "eval_accuracy": 0.6730769230769231, |
| "eval_loss": 0.9913409948348999, |
| "eval_runtime": 0.9733, |
| "eval_samples_per_second": 53.429, |
| "eval_steps_per_second": 1.027, |
| "step": 264 |
| }, |
| { |
| "epoch": 22.5, |
| "grad_norm": 11.394049644470215, |
| "learning_rate": 5.555555555555557e-06, |
| "loss": 0.3096, |
| "step": 270 |
| }, |
| { |
| "epoch": 23.0, |
| "eval_accuracy": 0.6346153846153846, |
| "eval_loss": 1.0354712009429932, |
| "eval_runtime": 0.9776, |
| "eval_samples_per_second": 53.194, |
| "eval_steps_per_second": 1.023, |
| "step": 276 |
| }, |
| { |
| "epoch": 23.333333333333332, |
| "grad_norm": 14.430404663085938, |
| "learning_rate": 4.938271604938272e-06, |
| "loss": 0.3603, |
| "step": 280 |
| }, |
| { |
| "epoch": 24.0, |
| "eval_accuracy": 0.6538461538461539, |
| "eval_loss": 0.998576819896698, |
| "eval_runtime": 1.0069, |
| "eval_samples_per_second": 51.645, |
| "eval_steps_per_second": 0.993, |
| "step": 288 |
| }, |
| { |
| "epoch": 24.166666666666668, |
| "grad_norm": 13.499276161193848, |
| "learning_rate": 4.3209876543209875e-06, |
| "loss": 0.3058, |
| "step": 290 |
| }, |
| { |
| "epoch": 25.0, |
| "grad_norm": 23.608278274536133, |
| "learning_rate": 3.7037037037037037e-06, |
| "loss": 0.2924, |
| "step": 300 |
| }, |
| { |
| "epoch": 25.0, |
| "eval_accuracy": 0.6730769230769231, |
| "eval_loss": 1.0046073198318481, |
| "eval_runtime": 1.3981, |
| "eval_samples_per_second": 37.194, |
| "eval_steps_per_second": 0.715, |
| "step": 300 |
| }, |
| { |
| "epoch": 25.833333333333332, |
| "grad_norm": 14.569175720214844, |
| "learning_rate": 3.08641975308642e-06, |
| "loss": 0.3489, |
| "step": 310 |
| }, |
| { |
| "epoch": 26.0, |
| "eval_accuracy": 0.6346153846153846, |
| "eval_loss": 1.0560189485549927, |
| "eval_runtime": 0.9923, |
| "eval_samples_per_second": 52.405, |
| "eval_steps_per_second": 1.008, |
| "step": 312 |
| }, |
| { |
| "epoch": 26.666666666666668, |
| "grad_norm": 8.643046379089355, |
| "learning_rate": 2.469135802469136e-06, |
| "loss": 0.2974, |
| "step": 320 |
| }, |
| { |
| "epoch": 27.0, |
| "eval_accuracy": 0.6538461538461539, |
| "eval_loss": 1.0075799226760864, |
| "eval_runtime": 1.2177, |
| "eval_samples_per_second": 42.703, |
| "eval_steps_per_second": 0.821, |
| "step": 324 |
| }, |
| { |
| "epoch": 27.5, |
| "grad_norm": 9.191235542297363, |
| "learning_rate": 1.8518518518518519e-06, |
| "loss": 0.2924, |
| "step": 330 |
| }, |
| { |
| "epoch": 28.0, |
| "eval_accuracy": 0.6538461538461539, |
| "eval_loss": 1.0164272785186768, |
| "eval_runtime": 1.0066, |
| "eval_samples_per_second": 51.659, |
| "eval_steps_per_second": 0.993, |
| "step": 336 |
| }, |
| { |
| "epoch": 28.333333333333332, |
| "grad_norm": 22.599172592163086, |
| "learning_rate": 1.234567901234568e-06, |
| "loss": 0.3369, |
| "step": 340 |
| }, |
| { |
| "epoch": 29.0, |
| "eval_accuracy": 0.6346153846153846, |
| "eval_loss": 1.0260354280471802, |
| "eval_runtime": 1.3444, |
| "eval_samples_per_second": 38.678, |
| "eval_steps_per_second": 0.744, |
| "step": 348 |
| }, |
| { |
| "epoch": 29.166666666666668, |
| "grad_norm": 11.389359474182129, |
| "learning_rate": 6.17283950617284e-07, |
| "loss": 0.3406, |
| "step": 350 |
| }, |
| { |
| "epoch": 30.0, |
| "grad_norm": 22.843719482421875, |
| "learning_rate": 0.0, |
| "loss": 0.2884, |
| "step": 360 |
| }, |
| { |
| "epoch": 30.0, |
| "eval_accuracy": 0.6346153846153846, |
| "eval_loss": 1.029270887374878, |
| "eval_runtime": 1.3179, |
| "eval_samples_per_second": 39.457, |
| "eval_steps_per_second": 0.759, |
| "step": 360 |
| }, |
| { |
| "epoch": 30.0, |
| "step": 360, |
| "total_flos": 7.115852825454182e+17, |
| "train_loss": 0.5890537910991245, |
| "train_runtime": 595.2576, |
| "train_samples_per_second": 36.74, |
| "train_steps_per_second": 0.605 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 360, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 30, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 7.115852825454182e+17, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
