| { | |
| "best_metric": 0.9338235294117647, | |
| "best_model_checkpoint": "vit-tiny-patch16-224-finetuned-papsmear/checkpoint-382", | |
| "epoch": 14.901960784313726, | |
| "eval_steps": 500, | |
| "global_step": 570, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.26143790849673204, | |
| "grad_norm": 22.882160186767578, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 1.9804, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5228758169934641, | |
| "grad_norm": 17.451801300048828, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 1.6798, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7843137254901961, | |
| "grad_norm": 14.190028190612793, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 1.4005, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.9934640522875817, | |
| "eval_accuracy": 0.5294117647058824, | |
| "eval_loss": 1.2214075326919556, | |
| "eval_runtime": 30.1025, | |
| "eval_samples_per_second": 4.518, | |
| "eval_steps_per_second": 0.565, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.0457516339869282, | |
| "grad_norm": 17.032344818115234, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 1.2065, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.3071895424836601, | |
| "grad_norm": 13.36404800415039, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 1.0837, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.5686274509803921, | |
| "grad_norm": 19.17522430419922, | |
| "learning_rate": 4.970760233918128e-05, | |
| "loss": 0.8728, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.8300653594771243, | |
| "grad_norm": 29.73708724975586, | |
| "learning_rate": 4.8732943469785574e-05, | |
| "loss": 0.8877, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.9869281045751634, | |
| "eval_accuracy": 0.6691176470588235, | |
| "eval_loss": 1.0726829767227173, | |
| "eval_runtime": 30.3082, | |
| "eval_samples_per_second": 4.487, | |
| "eval_steps_per_second": 0.561, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.0915032679738563, | |
| "grad_norm": 10.038992881774902, | |
| "learning_rate": 4.7758284600389865e-05, | |
| "loss": 0.8215, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 16.936914443969727, | |
| "learning_rate": 4.678362573099415e-05, | |
| "loss": 0.7058, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.6143790849673203, | |
| "grad_norm": 16.51180648803711, | |
| "learning_rate": 4.580896686159844e-05, | |
| "loss": 0.668, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.8758169934640523, | |
| "grad_norm": 14.309722900390625, | |
| "learning_rate": 4.483430799220273e-05, | |
| "loss": 0.603, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.980392156862745, | |
| "eval_accuracy": 0.7573529411764706, | |
| "eval_loss": 0.6806743144989014, | |
| "eval_runtime": 30.2645, | |
| "eval_samples_per_second": 4.494, | |
| "eval_steps_per_second": 0.562, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 3.1372549019607843, | |
| "grad_norm": 11.795894622802734, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 0.569, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.3986928104575163, | |
| "grad_norm": 13.41748046875, | |
| "learning_rate": 4.2884990253411305e-05, | |
| "loss": 0.4949, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.6601307189542482, | |
| "grad_norm": 18.057456970214844, | |
| "learning_rate": 4.1910331384015596e-05, | |
| "loss": 0.4527, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.9215686274509802, | |
| "grad_norm": 9.27607536315918, | |
| "learning_rate": 4.093567251461988e-05, | |
| "loss": 0.465, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7573529411764706, | |
| "eval_loss": 0.6485435366630554, | |
| "eval_runtime": 30.9055, | |
| "eval_samples_per_second": 4.401, | |
| "eval_steps_per_second": 0.55, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 4.183006535947713, | |
| "grad_norm": 15.063582420349121, | |
| "learning_rate": 3.996101364522417e-05, | |
| "loss": 0.4146, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "grad_norm": 9.970335960388184, | |
| "learning_rate": 3.898635477582846e-05, | |
| "loss": 0.3896, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 12.845563888549805, | |
| "learning_rate": 3.8011695906432746e-05, | |
| "loss": 0.4451, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.967320261437909, | |
| "grad_norm": 12.283594131469727, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.432, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.993464052287582, | |
| "eval_accuracy": 0.8014705882352942, | |
| "eval_loss": 0.5023567080497742, | |
| "eval_runtime": 30.7544, | |
| "eval_samples_per_second": 4.422, | |
| "eval_steps_per_second": 0.553, | |
| "step": 191 | |
| }, | |
| { | |
| "epoch": 5.228758169934641, | |
| "grad_norm": 10.823830604553223, | |
| "learning_rate": 3.606237816764133e-05, | |
| "loss": 0.3117, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.490196078431373, | |
| "grad_norm": 10.013704299926758, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 0.3033, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.751633986928105, | |
| "grad_norm": 4.759156227111816, | |
| "learning_rate": 3.41130604288499e-05, | |
| "loss": 0.2957, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.9869281045751634, | |
| "eval_accuracy": 0.8161764705882353, | |
| "eval_loss": 0.44854581356048584, | |
| "eval_runtime": 29.9105, | |
| "eval_samples_per_second": 4.547, | |
| "eval_steps_per_second": 0.568, | |
| "step": 229 | |
| }, | |
| { | |
| "epoch": 6.0130718954248366, | |
| "grad_norm": 13.833900451660156, | |
| "learning_rate": 3.313840155945419e-05, | |
| "loss": 0.3418, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 6.2745098039215685, | |
| "grad_norm": 10.218768119812012, | |
| "learning_rate": 3.216374269005848e-05, | |
| "loss": 0.2513, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 6.5359477124183005, | |
| "grad_norm": 8.130758285522461, | |
| "learning_rate": 3.118908382066277e-05, | |
| "loss": 0.2334, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.7973856209150325, | |
| "grad_norm": 11.395691871643066, | |
| "learning_rate": 3.0214424951267055e-05, | |
| "loss": 0.2203, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.980392156862745, | |
| "eval_accuracy": 0.8529411764705882, | |
| "eval_loss": 0.38499826192855835, | |
| "eval_runtime": 30.3064, | |
| "eval_samples_per_second": 4.488, | |
| "eval_steps_per_second": 0.561, | |
| "step": 267 | |
| }, | |
| { | |
| "epoch": 7.0588235294117645, | |
| "grad_norm": 27.90298080444336, | |
| "learning_rate": 2.9239766081871346e-05, | |
| "loss": 0.2809, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 7.3202614379084965, | |
| "grad_norm": 18.04613494873047, | |
| "learning_rate": 2.8265107212475634e-05, | |
| "loss": 0.3022, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 7.5816993464052285, | |
| "grad_norm": 13.312712669372559, | |
| "learning_rate": 2.729044834307992e-05, | |
| "loss": 0.2407, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 7.8431372549019605, | |
| "grad_norm": 6.36769962310791, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.236, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8455882352941176, | |
| "eval_loss": 0.36281630396842957, | |
| "eval_runtime": 29.9792, | |
| "eval_samples_per_second": 4.536, | |
| "eval_steps_per_second": 0.567, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 8.104575163398692, | |
| "grad_norm": 8.459989547729492, | |
| "learning_rate": 2.53411306042885e-05, | |
| "loss": 0.2048, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 8.366013071895425, | |
| "grad_norm": 10.514033317565918, | |
| "learning_rate": 2.4366471734892787e-05, | |
| "loss": 0.2115, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 8.627450980392156, | |
| "grad_norm": 5.906712055206299, | |
| "learning_rate": 2.3391812865497074e-05, | |
| "loss": 0.2652, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 8.88888888888889, | |
| "grad_norm": 10.485462188720703, | |
| "learning_rate": 2.2417153996101365e-05, | |
| "loss": 0.1857, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.993464052287582, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.29296010732650757, | |
| "eval_runtime": 30.4276, | |
| "eval_samples_per_second": 4.47, | |
| "eval_steps_per_second": 0.559, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 9.15032679738562, | |
| "grad_norm": 6.746099948883057, | |
| "learning_rate": 2.1442495126705653e-05, | |
| "loss": 0.1834, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 9.411764705882353, | |
| "grad_norm": 13.859869003295898, | |
| "learning_rate": 2.046783625730994e-05, | |
| "loss": 0.1163, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 9.673202614379084, | |
| "grad_norm": 11.039109230041504, | |
| "learning_rate": 1.949317738791423e-05, | |
| "loss": 0.1548, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 9.934640522875817, | |
| "grad_norm": 10.195306777954102, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.1907, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.986928104575163, | |
| "eval_accuracy": 0.9338235294117647, | |
| "eval_loss": 0.21208395063877106, | |
| "eval_runtime": 32.7787, | |
| "eval_samples_per_second": 4.149, | |
| "eval_steps_per_second": 0.519, | |
| "step": 382 | |
| }, | |
| { | |
| "epoch": 10.196078431372548, | |
| "grad_norm": 16.41983985900879, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 0.1645, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 10.457516339869281, | |
| "grad_norm": 13.68142032623291, | |
| "learning_rate": 1.6569200779727097e-05, | |
| "loss": 0.166, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 10.718954248366012, | |
| "grad_norm": 9.597548484802246, | |
| "learning_rate": 1.5594541910331384e-05, | |
| "loss": 0.1736, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 10.980392156862745, | |
| "grad_norm": 4.147403240203857, | |
| "learning_rate": 1.4619883040935673e-05, | |
| "loss": 0.1546, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 10.980392156862745, | |
| "eval_accuracy": 0.9264705882352942, | |
| "eval_loss": 0.2242465317249298, | |
| "eval_runtime": 31.8717, | |
| "eval_samples_per_second": 4.267, | |
| "eval_steps_per_second": 0.533, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 11.241830065359476, | |
| "grad_norm": 8.938773155212402, | |
| "learning_rate": 1.364522417153996e-05, | |
| "loss": 0.1202, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 11.50326797385621, | |
| "grad_norm": 5.6534199714660645, | |
| "learning_rate": 1.267056530214425e-05, | |
| "loss": 0.1263, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 11.764705882352942, | |
| "grad_norm": 11.322346687316895, | |
| "learning_rate": 1.1695906432748537e-05, | |
| "loss": 0.1375, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9191176470588235, | |
| "eval_loss": 0.19175946712493896, | |
| "eval_runtime": 30.8919, | |
| "eval_samples_per_second": 4.402, | |
| "eval_steps_per_second": 0.55, | |
| "step": 459 | |
| }, | |
| { | |
| "epoch": 12.026143790849673, | |
| "grad_norm": 3.5839948654174805, | |
| "learning_rate": 1.0721247563352826e-05, | |
| "loss": 0.0921, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 12.287581699346406, | |
| "grad_norm": 9.058156967163086, | |
| "learning_rate": 9.746588693957115e-06, | |
| "loss": 0.1056, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 12.549019607843137, | |
| "grad_norm": 18.891481399536133, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 0.1095, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 12.81045751633987, | |
| "grad_norm": 12.342869758605957, | |
| "learning_rate": 7.797270955165692e-06, | |
| "loss": 0.1237, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 12.993464052287582, | |
| "eval_accuracy": 0.9338235294117647, | |
| "eval_loss": 0.18091708421707153, | |
| "eval_runtime": 31.5611, | |
| "eval_samples_per_second": 4.309, | |
| "eval_steps_per_second": 0.539, | |
| "step": 497 | |
| }, | |
| { | |
| "epoch": 13.071895424836601, | |
| "grad_norm": 8.306052207946777, | |
| "learning_rate": 6.82261208576998e-06, | |
| "loss": 0.1274, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 13.333333333333334, | |
| "grad_norm": 9.888544082641602, | |
| "learning_rate": 5.8479532163742686e-06, | |
| "loss": 0.1125, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 13.594771241830065, | |
| "grad_norm": 12.163204193115234, | |
| "learning_rate": 4.873294346978558e-06, | |
| "loss": 0.0847, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 13.856209150326798, | |
| "grad_norm": 18.569528579711914, | |
| "learning_rate": 3.898635477582846e-06, | |
| "loss": 0.1637, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 13.986928104575163, | |
| "eval_accuracy": 0.9338235294117647, | |
| "eval_loss": 0.17736296355724335, | |
| "eval_runtime": 30.6036, | |
| "eval_samples_per_second": 4.444, | |
| "eval_steps_per_second": 0.555, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 14.117647058823529, | |
| "grad_norm": 9.332296371459961, | |
| "learning_rate": 2.9239766081871343e-06, | |
| "loss": 0.1376, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 14.379084967320262, | |
| "grad_norm": 8.473248481750488, | |
| "learning_rate": 1.949317738791423e-06, | |
| "loss": 0.1184, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 14.640522875816993, | |
| "grad_norm": 8.763771057128906, | |
| "learning_rate": 9.746588693957115e-07, | |
| "loss": 0.1036, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 14.901960784313726, | |
| "grad_norm": 8.387015342712402, | |
| "learning_rate": 0.0, | |
| "loss": 0.0803, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 14.901960784313726, | |
| "eval_accuracy": 0.9338235294117647, | |
| "eval_loss": 0.18818040192127228, | |
| "eval_runtime": 31.0653, | |
| "eval_samples_per_second": 4.378, | |
| "eval_steps_per_second": 0.547, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 14.901960784313726, | |
| "step": 570, | |
| "total_flos": 9.102711432019968e+16, | |
| "train_loss": 0.39481510582723117, | |
| "train_runtime": 4386.4067, | |
| "train_samples_per_second": 4.186, | |
| "train_steps_per_second": 0.13 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 570, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "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": 9.102711432019968e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |