{ "best_metric": 0.9297520661157025, "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-wsdmhar\\checkpoint-530", "epoch": 10.0, "eval_steps": 500, "global_step": 530, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.18867924528301888, "grad_norm": 5.205289840698242, "learning_rate": 9.433962264150944e-06, "loss": 1.7479, "step": 10 }, { "epoch": 0.37735849056603776, "grad_norm": 4.392179012298584, "learning_rate": 1.8867924528301888e-05, "loss": 1.4506, "step": 20 }, { "epoch": 0.5660377358490566, "grad_norm": 6.868739604949951, "learning_rate": 2.830188679245283e-05, "loss": 1.0283, "step": 30 }, { "epoch": 0.7547169811320755, "grad_norm": 9.729569435119629, "learning_rate": 3.7735849056603776e-05, "loss": 0.8064, "step": 40 }, { "epoch": 0.9433962264150944, "grad_norm": 12.286840438842773, "learning_rate": 4.716981132075472e-05, "loss": 0.7388, "step": 50 }, { "epoch": 1.0, "eval_accuracy": 0.7117768595041323, "eval_loss": 0.6308234930038452, "eval_runtime": 24.0663, "eval_samples_per_second": 120.667, "eval_steps_per_second": 3.781, "step": 53 }, { "epoch": 1.1320754716981132, "grad_norm": 9.05235481262207, "learning_rate": 4.9266247379454926e-05, "loss": 0.6616, "step": 60 }, { "epoch": 1.320754716981132, "grad_norm": 10.109884262084961, "learning_rate": 4.8218029350104823e-05, "loss": 0.6363, "step": 70 }, { "epoch": 1.509433962264151, "grad_norm": 10.793712615966797, "learning_rate": 4.716981132075472e-05, "loss": 0.554, "step": 80 }, { "epoch": 1.6981132075471699, "grad_norm": 17.802772521972656, "learning_rate": 4.612159329140461e-05, "loss": 0.5516, "step": 90 }, { "epoch": 1.8867924528301887, "grad_norm": 11.082262992858887, "learning_rate": 4.5073375262054504e-05, "loss": 0.5099, "step": 100 }, { "epoch": 2.0, "eval_accuracy": 0.8484848484848485, "eval_loss": 0.36690881848335266, "eval_runtime": 13.3175, "eval_samples_per_second": 218.059, "eval_steps_per_second": 6.833, "step": 106 }, { "epoch": 2.0754716981132075, "grad_norm": 6.183718204498291, "learning_rate": 4.402515723270441e-05, "loss": 0.4778, "step": 110 }, { "epoch": 2.2641509433962264, "grad_norm": 13.359389305114746, "learning_rate": 4.29769392033543e-05, "loss": 0.4449, "step": 120 }, { "epoch": 2.452830188679245, "grad_norm": 6.166327953338623, "learning_rate": 4.192872117400419e-05, "loss": 0.4243, "step": 130 }, { "epoch": 2.641509433962264, "grad_norm": 6.538959980010986, "learning_rate": 4.088050314465409e-05, "loss": 0.4265, "step": 140 }, { "epoch": 2.830188679245283, "grad_norm": 11.39398193359375, "learning_rate": 3.983228511530399e-05, "loss": 0.4319, "step": 150 }, { "epoch": 3.0, "eval_accuracy": 0.8684573002754821, "eval_loss": 0.33240532875061035, "eval_runtime": 14.2442, "eval_samples_per_second": 203.873, "eval_steps_per_second": 6.389, "step": 159 }, { "epoch": 3.018867924528302, "grad_norm": 9.0410795211792, "learning_rate": 3.878406708595388e-05, "loss": 0.3985, "step": 160 }, { "epoch": 3.207547169811321, "grad_norm": 6.786131381988525, "learning_rate": 3.7735849056603776e-05, "loss": 0.3912, "step": 170 }, { "epoch": 3.3962264150943398, "grad_norm": 8.235289573669434, "learning_rate": 3.6687631027253674e-05, "loss": 0.3496, "step": 180 }, { "epoch": 3.5849056603773586, "grad_norm": 6.866382122039795, "learning_rate": 3.5639412997903565e-05, "loss": 0.3696, "step": 190 }, { "epoch": 3.7735849056603774, "grad_norm": 8.17050552368164, "learning_rate": 3.4591194968553456e-05, "loss": 0.3203, "step": 200 }, { "epoch": 3.9622641509433962, "grad_norm": 17.23760414123535, "learning_rate": 3.354297693920336e-05, "loss": 0.4002, "step": 210 }, { "epoch": 4.0, "eval_accuracy": 0.9028925619834711, "eval_loss": 0.2758175730705261, "eval_runtime": 14.4978, "eval_samples_per_second": 200.306, "eval_steps_per_second": 6.277, "step": 212 }, { "epoch": 4.150943396226415, "grad_norm": 17.225961685180664, "learning_rate": 3.249475890985325e-05, "loss": 0.3484, "step": 220 }, { "epoch": 4.339622641509434, "grad_norm": 12.645174980163574, "learning_rate": 3.144654088050314e-05, "loss": 0.353, "step": 230 }, { "epoch": 4.528301886792453, "grad_norm": 5.725615978240967, "learning_rate": 3.0398322851153044e-05, "loss": 0.2883, "step": 240 }, { "epoch": 4.716981132075472, "grad_norm": 5.914320468902588, "learning_rate": 2.935010482180294e-05, "loss": 0.3067, "step": 250 }, { "epoch": 4.90566037735849, "grad_norm": 6.547543525695801, "learning_rate": 2.830188679245283e-05, "loss": 0.3589, "step": 260 }, { "epoch": 5.0, "eval_accuracy": 0.9132231404958677, "eval_loss": 0.2503372132778168, "eval_runtime": 14.406, "eval_samples_per_second": 201.582, "eval_steps_per_second": 6.317, "step": 265 }, { "epoch": 5.09433962264151, "grad_norm": 8.123590469360352, "learning_rate": 2.7253668763102725e-05, "loss": 0.3272, "step": 270 }, { "epoch": 5.283018867924528, "grad_norm": 5.585710048675537, "learning_rate": 2.6205450733752623e-05, "loss": 0.3535, "step": 280 }, { "epoch": 5.471698113207547, "grad_norm": 8.529882431030273, "learning_rate": 2.5157232704402517e-05, "loss": 0.2797, "step": 290 }, { "epoch": 5.660377358490566, "grad_norm": 6.009942054748535, "learning_rate": 2.4109014675052412e-05, "loss": 0.3004, "step": 300 }, { "epoch": 5.849056603773585, "grad_norm": 4.354741096496582, "learning_rate": 2.3060796645702306e-05, "loss": 0.3096, "step": 310 }, { "epoch": 6.0, "eval_accuracy": 0.9135674931129476, "eval_loss": 0.24190402030944824, "eval_runtime": 14.5262, "eval_samples_per_second": 199.915, "eval_steps_per_second": 6.265, "step": 318 }, { "epoch": 6.037735849056604, "grad_norm": 6.793937683105469, "learning_rate": 2.2012578616352204e-05, "loss": 0.3024, "step": 320 }, { "epoch": 6.226415094339623, "grad_norm": 4.6357221603393555, "learning_rate": 2.0964360587002095e-05, "loss": 0.3013, "step": 330 }, { "epoch": 6.415094339622642, "grad_norm": 5.533959865570068, "learning_rate": 1.9916142557651993e-05, "loss": 0.311, "step": 340 }, { "epoch": 6.60377358490566, "grad_norm": 5.0147600173950195, "learning_rate": 1.8867924528301888e-05, "loss": 0.2865, "step": 350 }, { "epoch": 6.7924528301886795, "grad_norm": 4.817899227142334, "learning_rate": 1.7819706498951782e-05, "loss": 0.2813, "step": 360 }, { "epoch": 6.981132075471698, "grad_norm": 6.032566547393799, "learning_rate": 1.677148846960168e-05, "loss": 0.2708, "step": 370 }, { "epoch": 7.0, "eval_accuracy": 0.9232093663911846, "eval_loss": 0.22773109376430511, "eval_runtime": 15.0479, "eval_samples_per_second": 192.984, "eval_steps_per_second": 6.047, "step": 371 }, { "epoch": 7.169811320754717, "grad_norm": 5.316783428192139, "learning_rate": 1.572327044025157e-05, "loss": 0.3009, "step": 380 }, { "epoch": 7.3584905660377355, "grad_norm": 5.2730607986450195, "learning_rate": 1.467505241090147e-05, "loss": 0.2422, "step": 390 }, { "epoch": 7.547169811320755, "grad_norm": 5.739044189453125, "learning_rate": 1.3626834381551362e-05, "loss": 0.2667, "step": 400 }, { "epoch": 7.735849056603773, "grad_norm": 5.689853668212891, "learning_rate": 1.2578616352201259e-05, "loss": 0.2833, "step": 410 }, { "epoch": 7.9245283018867925, "grad_norm": 6.43410062789917, "learning_rate": 1.1530398322851153e-05, "loss": 0.261, "step": 420 }, { "epoch": 8.0, "eval_accuracy": 0.925275482093664, "eval_loss": 0.21677668392658234, "eval_runtime": 14.7132, "eval_samples_per_second": 197.373, "eval_steps_per_second": 6.185, "step": 424 }, { "epoch": 8.11320754716981, "grad_norm": 5.5042643547058105, "learning_rate": 1.0482180293501048e-05, "loss": 0.2675, "step": 430 }, { "epoch": 8.30188679245283, "grad_norm": 4.880862236022949, "learning_rate": 9.433962264150944e-06, "loss": 0.2413, "step": 440 }, { "epoch": 8.49056603773585, "grad_norm": 3.9594244956970215, "learning_rate": 8.38574423480084e-06, "loss": 0.2577, "step": 450 }, { "epoch": 8.679245283018869, "grad_norm": 6.740973949432373, "learning_rate": 7.337526205450735e-06, "loss": 0.2595, "step": 460 }, { "epoch": 8.867924528301886, "grad_norm": 7.188548564910889, "learning_rate": 6.289308176100629e-06, "loss": 0.2526, "step": 470 }, { "epoch": 9.0, "eval_accuracy": 0.9245867768595041, "eval_loss": 0.20991356670856476, "eval_runtime": 14.793, "eval_samples_per_second": 196.31, "eval_steps_per_second": 6.152, "step": 477 }, { "epoch": 9.056603773584905, "grad_norm": 8.5458402633667, "learning_rate": 5.241090146750524e-06, "loss": 0.2465, "step": 480 }, { "epoch": 9.245283018867925, "grad_norm": 4.695093154907227, "learning_rate": 4.19287211740042e-06, "loss": 0.2172, "step": 490 }, { "epoch": 9.433962264150944, "grad_norm": 3.67899751663208, "learning_rate": 3.1446540880503146e-06, "loss": 0.2635, "step": 500 }, { "epoch": 9.622641509433961, "grad_norm": 4.685281276702881, "learning_rate": 2.09643605870021e-06, "loss": 0.254, "step": 510 }, { "epoch": 9.81132075471698, "grad_norm": 5.518146991729736, "learning_rate": 1.048218029350105e-06, "loss": 0.2574, "step": 520 }, { "epoch": 10.0, "grad_norm": 5.643453598022461, "learning_rate": 0.0, "loss": 0.2767, "step": 530 }, { "epoch": 10.0, "eval_accuracy": 0.9297520661157025, "eval_loss": 0.19896750152111053, "eval_runtime": 15.5471, "eval_samples_per_second": 186.787, "eval_steps_per_second": 5.853, "step": 530 }, { "epoch": 10.0, "step": 530, "total_flos": 1.6836842977571635e+18, "train_loss": 0.419760063009442, "train_runtime": 917.7632, "train_samples_per_second": 73.799, "train_steps_per_second": 0.577 } ], "logging_steps": 10, "max_steps": 530, "num_input_tokens_seen": 0, "num_train_epochs": 10, "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": 1.6836842977571635e+18, "train_batch_size": 32, "trial_name": null, "trial_params": null }