| { | |
| "best_metric": 0.9044117647058824, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-papsmear/checkpoint-497", | |
| "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": 12.742263793945312, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 1.84, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5228758169934641, | |
| "grad_norm": 12.05686092376709, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 1.6658, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7843137254901961, | |
| "grad_norm": 14.378918647766113, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 1.4844, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.9934640522875817, | |
| "eval_accuracy": 0.4117647058823529, | |
| "eval_loss": 1.429846167564392, | |
| "eval_runtime": 31.3549, | |
| "eval_samples_per_second": 4.337, | |
| "eval_steps_per_second": 0.542, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.0457516339869282, | |
| "grad_norm": 26.927827835083008, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 1.2683, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.3071895424836601, | |
| "grad_norm": 17.2440185546875, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 1.0979, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.5686274509803921, | |
| "grad_norm": 30.65620994567871, | |
| "learning_rate": 4.970760233918128e-05, | |
| "loss": 1.0747, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.8300653594771243, | |
| "grad_norm": 30.137895584106445, | |
| "learning_rate": 4.8732943469785574e-05, | |
| "loss": 0.8905, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.9869281045751634, | |
| "eval_accuracy": 0.6544117647058824, | |
| "eval_loss": 0.8109874129295349, | |
| "eval_runtime": 30.5109, | |
| "eval_samples_per_second": 4.457, | |
| "eval_steps_per_second": 0.557, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.0915032679738563, | |
| "grad_norm": 35.162410736083984, | |
| "learning_rate": 4.7758284600389865e-05, | |
| "loss": 0.7577, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 20.342853546142578, | |
| "learning_rate": 4.678362573099415e-05, | |
| "loss": 0.737, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.6143790849673203, | |
| "grad_norm": 20.470102310180664, | |
| "learning_rate": 4.580896686159844e-05, | |
| "loss": 0.7883, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.8758169934640523, | |
| "grad_norm": 46.66905212402344, | |
| "learning_rate": 4.483430799220273e-05, | |
| "loss": 0.9097, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.980392156862745, | |
| "eval_accuracy": 0.7132352941176471, | |
| "eval_loss": 0.7109317779541016, | |
| "eval_runtime": 31.3915, | |
| "eval_samples_per_second": 4.332, | |
| "eval_steps_per_second": 0.542, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 3.1372549019607843, | |
| "grad_norm": 20.6967716217041, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 0.6869, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.3986928104575163, | |
| "grad_norm": 17.672218322753906, | |
| "learning_rate": 4.2884990253411305e-05, | |
| "loss": 0.6266, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.6601307189542482, | |
| "grad_norm": 13.8719482421875, | |
| "learning_rate": 4.1910331384015596e-05, | |
| "loss": 0.521, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.9215686274509802, | |
| "grad_norm": 17.780046463012695, | |
| "learning_rate": 4.093567251461988e-05, | |
| "loss": 0.6238, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6544117647058824, | |
| "eval_loss": 0.9196979403495789, | |
| "eval_runtime": 30.0783, | |
| "eval_samples_per_second": 4.522, | |
| "eval_steps_per_second": 0.565, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 4.183006535947713, | |
| "grad_norm": 28.253620147705078, | |
| "learning_rate": 3.996101364522417e-05, | |
| "loss": 0.6683, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "grad_norm": 14.647994995117188, | |
| "learning_rate": 3.898635477582846e-05, | |
| "loss": 0.5111, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 21.987884521484375, | |
| "learning_rate": 3.8011695906432746e-05, | |
| "loss": 0.4989, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.967320261437909, | |
| "grad_norm": 14.68101692199707, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.4456, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.993464052287582, | |
| "eval_accuracy": 0.8014705882352942, | |
| "eval_loss": 0.46520400047302246, | |
| "eval_runtime": 29.9903, | |
| "eval_samples_per_second": 4.535, | |
| "eval_steps_per_second": 0.567, | |
| "step": 191 | |
| }, | |
| { | |
| "epoch": 5.228758169934641, | |
| "grad_norm": 10.403129577636719, | |
| "learning_rate": 3.606237816764133e-05, | |
| "loss": 0.3867, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.490196078431373, | |
| "grad_norm": 22.698213577270508, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 0.3797, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.751633986928105, | |
| "grad_norm": 15.229155540466309, | |
| "learning_rate": 3.41130604288499e-05, | |
| "loss": 0.4394, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.9869281045751634, | |
| "eval_accuracy": 0.8014705882352942, | |
| "eval_loss": 0.5187686681747437, | |
| "eval_runtime": 31.0968, | |
| "eval_samples_per_second": 4.373, | |
| "eval_steps_per_second": 0.547, | |
| "step": 229 | |
| }, | |
| { | |
| "epoch": 6.0130718954248366, | |
| "grad_norm": 15.33947467803955, | |
| "learning_rate": 3.313840155945419e-05, | |
| "loss": 0.3482, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 6.2745098039215685, | |
| "grad_norm": 27.592214584350586, | |
| "learning_rate": 3.216374269005848e-05, | |
| "loss": 0.3717, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 6.5359477124183005, | |
| "grad_norm": 16.805540084838867, | |
| "learning_rate": 3.118908382066277e-05, | |
| "loss": 0.3361, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.7973856209150325, | |
| "grad_norm": 19.967639923095703, | |
| "learning_rate": 3.0214424951267055e-05, | |
| "loss": 0.3156, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.980392156862745, | |
| "eval_accuracy": 0.8529411764705882, | |
| "eval_loss": 0.34466856718063354, | |
| "eval_runtime": 30.7808, | |
| "eval_samples_per_second": 4.418, | |
| "eval_steps_per_second": 0.552, | |
| "step": 267 | |
| }, | |
| { | |
| "epoch": 7.0588235294117645, | |
| "grad_norm": 15.27043628692627, | |
| "learning_rate": 2.9239766081871346e-05, | |
| "loss": 0.284, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 7.3202614379084965, | |
| "grad_norm": 20.704370498657227, | |
| "learning_rate": 2.8265107212475634e-05, | |
| "loss": 0.2777, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 7.5816993464052285, | |
| "grad_norm": 19.92112922668457, | |
| "learning_rate": 2.729044834307992e-05, | |
| "loss": 0.2911, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 7.8431372549019605, | |
| "grad_norm": 23.951862335205078, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.2212, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8382352941176471, | |
| "eval_loss": 0.3509288728237152, | |
| "eval_runtime": 30.877, | |
| "eval_samples_per_second": 4.405, | |
| "eval_steps_per_second": 0.551, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 8.104575163398692, | |
| "grad_norm": 28.307418823242188, | |
| "learning_rate": 2.53411306042885e-05, | |
| "loss": 0.221, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 8.366013071895425, | |
| "grad_norm": 12.33401107788086, | |
| "learning_rate": 2.4366471734892787e-05, | |
| "loss": 0.1945, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 8.627450980392156, | |
| "grad_norm": 26.391366958618164, | |
| "learning_rate": 2.3391812865497074e-05, | |
| "loss": 0.2758, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 8.88888888888889, | |
| "grad_norm": 10.806180000305176, | |
| "learning_rate": 2.2417153996101365e-05, | |
| "loss": 0.2402, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.993464052287582, | |
| "eval_accuracy": 0.8235294117647058, | |
| "eval_loss": 0.393919974565506, | |
| "eval_runtime": 31.1996, | |
| "eval_samples_per_second": 4.359, | |
| "eval_steps_per_second": 0.545, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 9.15032679738562, | |
| "grad_norm": 16.904836654663086, | |
| "learning_rate": 2.1442495126705653e-05, | |
| "loss": 0.1835, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 9.411764705882353, | |
| "grad_norm": 30.91939926147461, | |
| "learning_rate": 2.046783625730994e-05, | |
| "loss": 0.1562, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 9.673202614379084, | |
| "grad_norm": 14.6829833984375, | |
| "learning_rate": 1.949317738791423e-05, | |
| "loss": 0.195, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 9.934640522875817, | |
| "grad_norm": 26.035619735717773, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.1733, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.986928104575163, | |
| "eval_accuracy": 0.8897058823529411, | |
| "eval_loss": 0.24441416561603546, | |
| "eval_runtime": 30.8179, | |
| "eval_samples_per_second": 4.413, | |
| "eval_steps_per_second": 0.552, | |
| "step": 382 | |
| }, | |
| { | |
| "epoch": 10.196078431372548, | |
| "grad_norm": 14.318477630615234, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 0.1458, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 10.457516339869281, | |
| "grad_norm": 9.527152061462402, | |
| "learning_rate": 1.6569200779727097e-05, | |
| "loss": 0.1752, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 10.718954248366012, | |
| "grad_norm": 22.481767654418945, | |
| "learning_rate": 1.5594541910331384e-05, | |
| "loss": 0.1969, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 10.980392156862745, | |
| "grad_norm": 42.72972869873047, | |
| "learning_rate": 1.4619883040935673e-05, | |
| "loss": 0.1953, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 10.980392156862745, | |
| "eval_accuracy": 0.8676470588235294, | |
| "eval_loss": 0.26393771171569824, | |
| "eval_runtime": 31.9368, | |
| "eval_samples_per_second": 4.258, | |
| "eval_steps_per_second": 0.532, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 11.241830065359476, | |
| "grad_norm": 10.306249618530273, | |
| "learning_rate": 1.364522417153996e-05, | |
| "loss": 0.1626, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 11.50326797385621, | |
| "grad_norm": 9.890297889709473, | |
| "learning_rate": 1.267056530214425e-05, | |
| "loss": 0.1885, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 11.764705882352942, | |
| "grad_norm": 13.813959121704102, | |
| "learning_rate": 1.1695906432748537e-05, | |
| "loss": 0.1363, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8823529411764706, | |
| "eval_loss": 0.264517605304718, | |
| "eval_runtime": 30.4905, | |
| "eval_samples_per_second": 4.46, | |
| "eval_steps_per_second": 0.558, | |
| "step": 459 | |
| }, | |
| { | |
| "epoch": 12.026143790849673, | |
| "grad_norm": 8.809865951538086, | |
| "learning_rate": 1.0721247563352826e-05, | |
| "loss": 0.0897, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 12.287581699346406, | |
| "grad_norm": 9.379281997680664, | |
| "learning_rate": 9.746588693957115e-06, | |
| "loss": 0.0972, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 12.549019607843137, | |
| "grad_norm": 42.292972564697266, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 0.1417, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 12.81045751633987, | |
| "grad_norm": 16.6181697845459, | |
| "learning_rate": 7.797270955165692e-06, | |
| "loss": 0.1234, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 12.993464052287582, | |
| "eval_accuracy": 0.9044117647058824, | |
| "eval_loss": 0.20269162952899933, | |
| "eval_runtime": 30.5106, | |
| "eval_samples_per_second": 4.457, | |
| "eval_steps_per_second": 0.557, | |
| "step": 497 | |
| }, | |
| { | |
| "epoch": 13.071895424836601, | |
| "grad_norm": 28.557069778442383, | |
| "learning_rate": 6.82261208576998e-06, | |
| "loss": 0.1402, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 13.333333333333334, | |
| "grad_norm": 22.95408058166504, | |
| "learning_rate": 5.8479532163742686e-06, | |
| "loss": 0.1396, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 13.594771241830065, | |
| "grad_norm": 17.663022994995117, | |
| "learning_rate": 4.873294346978558e-06, | |
| "loss": 0.0941, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 13.856209150326798, | |
| "grad_norm": 37.5713996887207, | |
| "learning_rate": 3.898635477582846e-06, | |
| "loss": 0.1282, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 13.986928104575163, | |
| "eval_accuracy": 0.9044117647058824, | |
| "eval_loss": 0.20270010828971863, | |
| "eval_runtime": 30.5276, | |
| "eval_samples_per_second": 4.455, | |
| "eval_steps_per_second": 0.557, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 14.117647058823529, | |
| "grad_norm": 15.502400398254395, | |
| "learning_rate": 2.9239766081871343e-06, | |
| "loss": 0.1222, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 14.379084967320262, | |
| "grad_norm": 30.627925872802734, | |
| "learning_rate": 1.949317738791423e-06, | |
| "loss": 0.1129, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 14.640522875816993, | |
| "grad_norm": 19.545286178588867, | |
| "learning_rate": 9.746588693957115e-07, | |
| "loss": 0.1122, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 14.901960784313726, | |
| "grad_norm": 8.540277481079102, | |
| "learning_rate": 0.0, | |
| "loss": 0.1036, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 14.901960784313726, | |
| "eval_accuracy": 0.8897058823529411, | |
| "eval_loss": 0.20258459448814392, | |
| "eval_runtime": 30.3873, | |
| "eval_samples_per_second": 4.476, | |
| "eval_steps_per_second": 0.559, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 14.901960784313726, | |
| "step": 570, | |
| "total_flos": 4.5342391246258176e+17, | |
| "train_loss": 0.4384694022044801, | |
| "train_runtime": 4524.6194, | |
| "train_samples_per_second": 4.058, | |
| "train_steps_per_second": 0.126 | |
| } | |
| ], | |
| "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": 4.5342391246258176e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |