| { | |
| "best_metric": 0.37125059962272644, | |
| "best_model_checkpoint": "Model-SwinV2-Large-\\checkpoint-407", | |
| "epoch": 10.980132450331126, | |
| "eval_steps": 7, | |
| "global_step": 407, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18543046357615894, | |
| "grad_norm": 7.775666236877441, | |
| "learning_rate": 8.108108108108109e-07, | |
| "loss": 5.4649, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.3708609271523179, | |
| "grad_norm": 9.706766128540039, | |
| "learning_rate": 1.756756756756757e-06, | |
| "loss": 5.4623, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5562913907284768, | |
| "grad_norm": 7.272828578948975, | |
| "learning_rate": 2.702702702702703e-06, | |
| "loss": 5.4689, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.7417218543046358, | |
| "grad_norm": 9.961599349975586, | |
| "learning_rate": 3.648648648648649e-06, | |
| "loss": 5.3771, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.9271523178807947, | |
| "grad_norm": 9.451801300048828, | |
| "learning_rate": 4.594594594594595e-06, | |
| "loss": 5.3897, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.9801324503311258, | |
| "eval_accuracy": 0.009016393442622951, | |
| "eval_f1_macro": 0.008969822352444303, | |
| "eval_f1_micro": 0.009016393442622951, | |
| "eval_f1_weighted": 0.008562213642681454, | |
| "eval_loss": 5.330114841461182, | |
| "eval_precision_macro": 0.012256033182503772, | |
| "eval_precision_micro": 0.009016393442622951, | |
| "eval_precision_weighted": 0.011259550478451154, | |
| "eval_recall_macro": 0.009261904761904761, | |
| "eval_recall_micro": 0.009016393442622951, | |
| "eval_recall_weighted": 0.009016393442622951, | |
| "eval_runtime": 84.595, | |
| "eval_samples_per_second": 14.422, | |
| "eval_steps_per_second": 0.236, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.1324503311258278, | |
| "grad_norm": 8.083910942077637, | |
| "learning_rate": 5.5405405405405415e-06, | |
| "loss": 5.5232, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.3178807947019868, | |
| "grad_norm": 16.997234344482422, | |
| "learning_rate": 6.486486486486487e-06, | |
| "loss": 5.3019, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.5033112582781456, | |
| "grad_norm": 8.079388618469238, | |
| "learning_rate": 7.297297297297298e-06, | |
| "loss": 5.2404, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.6887417218543046, | |
| "grad_norm": 7.1500725746154785, | |
| "learning_rate": 8.243243243243245e-06, | |
| "loss": 5.218, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.8741721854304636, | |
| "grad_norm": 6.12971830368042, | |
| "learning_rate": 9.18918918918919e-06, | |
| "loss": 5.1645, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.980132450331126, | |
| "eval_accuracy": 0.05245901639344262, | |
| "eval_f1_macro": 0.03662301941302143, | |
| "eval_f1_micro": 0.05245901639344262, | |
| "eval_f1_weighted": 0.0388490193249229, | |
| "eval_loss": 5.068942070007324, | |
| "eval_precision_macro": 0.03409158543573505, | |
| "eval_precision_micro": 0.05245901639344262, | |
| "eval_precision_weighted": 0.035601775905091275, | |
| "eval_recall_macro": 0.04867261904761905, | |
| "eval_recall_micro": 0.05245901639344262, | |
| "eval_recall_weighted": 0.05245901639344262, | |
| "eval_runtime": 60.2034, | |
| "eval_samples_per_second": 20.265, | |
| "eval_steps_per_second": 0.332, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.0794701986754967, | |
| "grad_norm": 10.855502128601074, | |
| "learning_rate": 1.0135135135135136e-05, | |
| "loss": 5.2745, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.2649006622516556, | |
| "grad_norm": 8.226344108581543, | |
| "learning_rate": 1.0945945945945948e-05, | |
| "loss": 5.0241, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.4503311258278146, | |
| "grad_norm": 6.742478370666504, | |
| "learning_rate": 1.1891891891891893e-05, | |
| "loss": 4.9665, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 2.6357615894039736, | |
| "grad_norm": 8.76828384399414, | |
| "learning_rate": 1.2837837837837838e-05, | |
| "loss": 4.7954, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 2.821192052980132, | |
| "grad_norm": 16.411344528198242, | |
| "learning_rate": 1.3783783783783786e-05, | |
| "loss": 4.6041, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.980132450331126, | |
| "eval_accuracy": 0.26311475409836066, | |
| "eval_f1_macro": 0.20009963945415252, | |
| "eval_f1_micro": 0.26311475409836066, | |
| "eval_f1_weighted": 0.21148759565581807, | |
| "eval_loss": 3.9028923511505127, | |
| "eval_precision_macro": 0.22816811623497163, | |
| "eval_precision_micro": 0.26311475409836066, | |
| "eval_precision_weighted": 0.23705714211336962, | |
| "eval_recall_macro": 0.24582738095238096, | |
| "eval_recall_micro": 0.26311475409836066, | |
| "eval_recall_weighted": 0.26311475409836066, | |
| "eval_runtime": 110.6815, | |
| "eval_samples_per_second": 11.023, | |
| "eval_steps_per_second": 0.181, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 3.0264900662251657, | |
| "grad_norm": 14.454651832580566, | |
| "learning_rate": 1.472972972972973e-05, | |
| "loss": 4.4593, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 3.2119205298013247, | |
| "grad_norm": 33.457191467285156, | |
| "learning_rate": 1.5675675675675676e-05, | |
| "loss": 3.7871, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 3.3973509933774833, | |
| "grad_norm": 20.33750343322754, | |
| "learning_rate": 1.662162162162162e-05, | |
| "loss": 3.3659, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 3.5827814569536423, | |
| "grad_norm": 17.52394676208496, | |
| "learning_rate": 1.7432432432432434e-05, | |
| "loss": 2.9398, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 3.7682119205298013, | |
| "grad_norm": 38.1245002746582, | |
| "learning_rate": 1.837837837837838e-05, | |
| "loss": 2.5717, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.9536423841059603, | |
| "grad_norm": 40.0474967956543, | |
| "learning_rate": 1.9324324324324324e-05, | |
| "loss": 2.0853, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 3.980132450331126, | |
| "eval_accuracy": 0.6860655737704918, | |
| "eval_f1_macro": 0.6476607878873493, | |
| "eval_f1_micro": 0.6860655737704918, | |
| "eval_f1_weighted": 0.6600757648502281, | |
| "eval_loss": 1.2792028188705444, | |
| "eval_precision_macro": 0.6980460318405365, | |
| "eval_precision_micro": 0.6860655737704918, | |
| "eval_precision_weighted": 0.7044202635761663, | |
| "eval_recall_macro": 0.6700654761904761, | |
| "eval_recall_micro": 0.6860655737704918, | |
| "eval_recall_weighted": 0.6860655737704918, | |
| "eval_runtime": 47.8473, | |
| "eval_samples_per_second": 25.498, | |
| "eval_steps_per_second": 0.418, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 4.158940397350993, | |
| "grad_norm": 14.780089378356934, | |
| "learning_rate": 2.0270270270270273e-05, | |
| "loss": 1.7539, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 4.344370860927152, | |
| "grad_norm": 23.0028076171875, | |
| "learning_rate": 2.1216216216216218e-05, | |
| "loss": 1.4574, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 4.529801324503311, | |
| "grad_norm": 14.302562713623047, | |
| "learning_rate": 2.2162162162162166e-05, | |
| "loss": 1.3244, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 4.71523178807947, | |
| "grad_norm": 15.798663139343262, | |
| "learning_rate": 2.3108108108108108e-05, | |
| "loss": 1.2262, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 4.900662251655629, | |
| "grad_norm": 18.089494705200195, | |
| "learning_rate": 2.4054054054054056e-05, | |
| "loss": 1.2735, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 4.9801324503311255, | |
| "eval_accuracy": 0.8327868852459016, | |
| "eval_f1_macro": 0.8214352639534529, | |
| "eval_f1_micro": 0.8327868852459016, | |
| "eval_f1_weighted": 0.8270994593768409, | |
| "eval_loss": 0.6065749526023865, | |
| "eval_precision_macro": 0.8606998556998556, | |
| "eval_precision_micro": 0.8327868852459016, | |
| "eval_precision_weighted": 0.8636802746433894, | |
| "eval_recall_macro": 0.8253214285714285, | |
| "eval_recall_micro": 0.8327868852459016, | |
| "eval_recall_weighted": 0.8327868852459016, | |
| "eval_runtime": 83.6381, | |
| "eval_samples_per_second": 14.587, | |
| "eval_steps_per_second": 0.239, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 5.105960264900662, | |
| "grad_norm": 16.889223098754883, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.0632, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 5.291390728476821, | |
| "grad_norm": 12.659591674804688, | |
| "learning_rate": 2.594594594594595e-05, | |
| "loss": 0.9143, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 5.47682119205298, | |
| "grad_norm": 10.869107246398926, | |
| "learning_rate": 2.689189189189189e-05, | |
| "loss": 0.9349, | |
| "step": 203 | |
| }, | |
| { | |
| "epoch": 5.662251655629139, | |
| "grad_norm": 15.735921859741211, | |
| "learning_rate": 2.7837837837837836e-05, | |
| "loss": 0.8886, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.847682119205298, | |
| "grad_norm": 13.42103385925293, | |
| "learning_rate": 2.8783783783783785e-05, | |
| "loss": 0.8957, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 5.9801324503311255, | |
| "eval_accuracy": 0.8754098360655738, | |
| "eval_f1_macro": 0.8704101537007883, | |
| "eval_f1_micro": 0.8754098360655738, | |
| "eval_f1_weighted": 0.8727622563744599, | |
| "eval_loss": 0.4349415898323059, | |
| "eval_precision_macro": 0.8963481005709266, | |
| "eval_precision_micro": 0.8754098360655738, | |
| "eval_precision_weighted": 0.8973616240811536, | |
| "eval_recall_macro": 0.8722559523809523, | |
| "eval_recall_micro": 0.8754098360655738, | |
| "eval_recall_weighted": 0.8754098360655738, | |
| "eval_runtime": 86.4338, | |
| "eval_samples_per_second": 14.115, | |
| "eval_steps_per_second": 0.231, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 6.052980132450331, | |
| "grad_norm": 13.915326118469238, | |
| "learning_rate": 2.9729729729729733e-05, | |
| "loss": 0.8343, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 6.23841059602649, | |
| "grad_norm": 11.141881942749023, | |
| "learning_rate": 3.0675675675675675e-05, | |
| "loss": 0.742, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 6.423841059602649, | |
| "grad_norm": 13.977461814880371, | |
| "learning_rate": 3.162162162162162e-05, | |
| "loss": 0.7757, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 6.609271523178808, | |
| "grad_norm": 12.792037010192871, | |
| "learning_rate": 3.256756756756757e-05, | |
| "loss": 0.7528, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 6.7947019867549665, | |
| "grad_norm": 9.550854682922363, | |
| "learning_rate": 3.351351351351352e-05, | |
| "loss": 0.7589, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 6.9801324503311255, | |
| "grad_norm": 56.45753860473633, | |
| "learning_rate": 3.445945945945946e-05, | |
| "loss": 0.7522, | |
| "step": 259 | |
| }, | |
| { | |
| "epoch": 6.9801324503311255, | |
| "eval_accuracy": 0.8885245901639345, | |
| "eval_f1_macro": 0.8810641592669457, | |
| "eval_f1_micro": 0.8885245901639345, | |
| "eval_f1_weighted": 0.8837497312389461, | |
| "eval_loss": 0.41575273871421814, | |
| "eval_precision_macro": 0.9128655927405928, | |
| "eval_precision_micro": 0.8885245901639345, | |
| "eval_precision_weighted": 0.9129847338453896, | |
| "eval_recall_macro": 0.8841249999999999, | |
| "eval_recall_micro": 0.8885245901639345, | |
| "eval_recall_weighted": 0.8885245901639345, | |
| "eval_runtime": 87.2308, | |
| "eval_samples_per_second": 13.986, | |
| "eval_steps_per_second": 0.229, | |
| "step": 259 | |
| }, | |
| { | |
| "epoch": 7.185430463576159, | |
| "grad_norm": 9.251741409301758, | |
| "learning_rate": 3.5405405405405403e-05, | |
| "loss": 0.71, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 7.370860927152318, | |
| "grad_norm": 12.447789192199707, | |
| "learning_rate": 3.635135135135135e-05, | |
| "loss": 0.6271, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 7.556291390728477, | |
| "grad_norm": 15.672873497009277, | |
| "learning_rate": 3.7162162162162165e-05, | |
| "loss": 0.704, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 7.741721854304636, | |
| "grad_norm": 21.882080078125, | |
| "learning_rate": 3.8108108108108106e-05, | |
| "loss": 0.6397, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 7.927152317880795, | |
| "grad_norm": 12.572905540466309, | |
| "learning_rate": 3.9054054054054055e-05, | |
| "loss": 0.6157, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 7.9801324503311255, | |
| "eval_accuracy": 0.8860655737704918, | |
| "eval_f1_macro": 0.8812874675633964, | |
| "eval_f1_micro": 0.8860655737704918, | |
| "eval_f1_weighted": 0.883605155706993, | |
| "eval_loss": 0.37819206714630127, | |
| "eval_precision_macro": 0.9077945526695527, | |
| "eval_precision_micro": 0.8860655737704918, | |
| "eval_precision_weighted": 0.9074083043550256, | |
| "eval_recall_macro": 0.8824166666666667, | |
| "eval_recall_micro": 0.8860655737704918, | |
| "eval_recall_weighted": 0.8860655737704918, | |
| "eval_runtime": 86.9903, | |
| "eval_samples_per_second": 14.025, | |
| "eval_steps_per_second": 0.23, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 8.132450331125828, | |
| "grad_norm": 9.28415298461914, | |
| "learning_rate": 4e-05, | |
| "loss": 0.632, | |
| "step": 301 | |
| }, | |
| { | |
| "epoch": 8.317880794701987, | |
| "grad_norm": 9.938474655151367, | |
| "learning_rate": 4.0945945945945945e-05, | |
| "loss": 0.5854, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 8.503311258278146, | |
| "grad_norm": 9.133801460266113, | |
| "learning_rate": 4.189189189189189e-05, | |
| "loss": 0.4822, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 8.688741721854305, | |
| "grad_norm": 9.311871528625488, | |
| "learning_rate": 4.283783783783784e-05, | |
| "loss": 0.5033, | |
| "step": 322 | |
| }, | |
| { | |
| "epoch": 8.874172185430464, | |
| "grad_norm": 8.707598686218262, | |
| "learning_rate": 4.378378378378379e-05, | |
| "loss": 0.6243, | |
| "step": 329 | |
| }, | |
| { | |
| "epoch": 8.980132450331126, | |
| "eval_accuracy": 0.8795081967213115, | |
| "eval_f1_macro": 0.8727452907859256, | |
| "eval_f1_micro": 0.8795081967213115, | |
| "eval_f1_weighted": 0.8751382461299096, | |
| "eval_loss": 0.4089548587799072, | |
| "eval_precision_macro": 0.8947101648351647, | |
| "eval_precision_micro": 0.8795081967213115, | |
| "eval_precision_weighted": 0.8962854563469319, | |
| "eval_recall_macro": 0.8770833333333334, | |
| "eval_recall_micro": 0.8795081967213115, | |
| "eval_recall_weighted": 0.8795081967213115, | |
| "eval_runtime": 85.9321, | |
| "eval_samples_per_second": 14.197, | |
| "eval_steps_per_second": 0.233, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 9.079470198675496, | |
| "grad_norm": 10.794309616088867, | |
| "learning_rate": 4.472972972972973e-05, | |
| "loss": 0.5572, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 9.264900662251655, | |
| "grad_norm": 10.890398979187012, | |
| "learning_rate": 4.567567567567568e-05, | |
| "loss": 0.5487, | |
| "step": 343 | |
| }, | |
| { | |
| "epoch": 9.450331125827814, | |
| "grad_norm": 10.409049987792969, | |
| "learning_rate": 4.662162162162162e-05, | |
| "loss": 0.6332, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 9.635761589403973, | |
| "grad_norm": 9.35953426361084, | |
| "learning_rate": 4.756756756756757e-05, | |
| "loss": 0.5324, | |
| "step": 357 | |
| }, | |
| { | |
| "epoch": 9.821192052980132, | |
| "grad_norm": 6.873753547668457, | |
| "learning_rate": 4.851351351351351e-05, | |
| "loss": 0.539, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 9.980132450331126, | |
| "eval_accuracy": 0.8811475409836066, | |
| "eval_f1_macro": 0.8763333539488136, | |
| "eval_f1_micro": 0.8811475409836066, | |
| "eval_f1_weighted": 0.878023237317979, | |
| "eval_loss": 0.39534232020378113, | |
| "eval_precision_macro": 0.9026571067821066, | |
| "eval_precision_micro": 0.8811475409836066, | |
| "eval_precision_weighted": 0.9016379544863151, | |
| "eval_recall_macro": 0.8773511904761905, | |
| "eval_recall_micro": 0.8811475409836066, | |
| "eval_recall_weighted": 0.8811475409836066, | |
| "eval_runtime": 87.5111, | |
| "eval_samples_per_second": 13.941, | |
| "eval_steps_per_second": 0.229, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 10.026490066225165, | |
| "grad_norm": 13.179052352905273, | |
| "learning_rate": 4.945945945945946e-05, | |
| "loss": 0.5554, | |
| "step": 371 | |
| }, | |
| { | |
| "epoch": 10.211920529801324, | |
| "grad_norm": 7.148564338684082, | |
| "learning_rate": 4.995495495495496e-05, | |
| "loss": 0.4887, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 10.397350993377483, | |
| "grad_norm": 9.368721961975098, | |
| "learning_rate": 4.984984984984985e-05, | |
| "loss": 0.5459, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 10.582781456953642, | |
| "grad_norm": 10.710000038146973, | |
| "learning_rate": 4.9744744744744746e-05, | |
| "loss": 0.5324, | |
| "step": 392 | |
| }, | |
| { | |
| "epoch": 10.7682119205298, | |
| "grad_norm": 11.22227668762207, | |
| "learning_rate": 4.963963963963964e-05, | |
| "loss": 0.5425, | |
| "step": 399 | |
| }, | |
| { | |
| "epoch": 10.95364238410596, | |
| "grad_norm": 8.322479248046875, | |
| "learning_rate": 4.953453453453454e-05, | |
| "loss": 0.5839, | |
| "step": 406 | |
| }, | |
| { | |
| "epoch": 10.980132450331126, | |
| "eval_accuracy": 0.8918032786885246, | |
| "eval_f1_macro": 0.8875326968782852, | |
| "eval_f1_micro": 0.8918032786885246, | |
| "eval_f1_weighted": 0.889380159243949, | |
| "eval_loss": 0.37125059962272644, | |
| "eval_precision_macro": 0.9082038517038518, | |
| "eval_precision_micro": 0.8918032786885246, | |
| "eval_precision_weighted": 0.9081954816585965, | |
| "eval_recall_macro": 0.8885952380952381, | |
| "eval_recall_micro": 0.8918032786885246, | |
| "eval_recall_weighted": 0.8918032786885246, | |
| "eval_runtime": 130.2164, | |
| "eval_samples_per_second": 9.369, | |
| "eval_steps_per_second": 0.154, | |
| "step": 407 | |
| } | |
| ], | |
| "logging_steps": 7, | |
| "max_steps": 3700, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 7, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.01 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 3 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.804145744376758e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |