| { | |
| "best_metric": 0.6718301778542742, | |
| "best_model_checkpoint": "vit-base-patch16-224-best-finetuned-on-affectnet_short/checkpoint-672", | |
| "epoch": 22.0, | |
| "global_step": 704, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 7.042253521126762e-06, | |
| "loss": 2.1884, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.4084507042253523e-05, | |
| "loss": 2.0969, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 2.112676056338028e-05, | |
| "loss": 1.9968, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.27538726333907054, | |
| "eval_f1": 0.22796602838082472, | |
| "eval_loss": 1.9113303422927856, | |
| "eval_precision": 0.2518389295969545, | |
| "eval_recall": 0.27538726333907054, | |
| "eval_runtime": 25.038, | |
| "eval_samples_per_second": 69.614, | |
| "eval_steps_per_second": 4.353, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 2.8169014084507046e-05, | |
| "loss": 1.8758, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 3.5211267605633805e-05, | |
| "loss": 1.6422, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 4.225352112676056e-05, | |
| "loss": 1.4178, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5048766494549627, | |
| "eval_f1": 0.4900447126711682, | |
| "eval_loss": 1.2704347372055054, | |
| "eval_precision": 0.5149366289457634, | |
| "eval_recall": 0.5048766494549627, | |
| "eval_runtime": 24.2957, | |
| "eval_samples_per_second": 71.741, | |
| "eval_steps_per_second": 4.486, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 4.929577464788733e-05, | |
| "loss": 1.2352, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 4.928909952606635e-05, | |
| "loss": 1.1754, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 4.849921011058452e-05, | |
| "loss": 1.1751, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5840504876649455, | |
| "eval_f1": 0.5787373156833001, | |
| "eval_loss": 1.111608624458313, | |
| "eval_precision": 0.5891362431175351, | |
| "eval_recall": 0.5840504876649455, | |
| "eval_runtime": 24.5236, | |
| "eval_samples_per_second": 71.074, | |
| "eval_steps_per_second": 4.445, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 4.770932069510269e-05, | |
| "loss": 1.1688, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 4.691943127962086e-05, | |
| "loss": 1.0893, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 4.6129541864139027e-05, | |
| "loss": 1.0127, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6161790017211703, | |
| "eval_f1": 0.6141128202096976, | |
| "eval_loss": 1.0236890316009521, | |
| "eval_precision": 0.633451007227001, | |
| "eval_recall": 0.6161790017211703, | |
| "eval_runtime": 24.3337, | |
| "eval_samples_per_second": 71.629, | |
| "eval_steps_per_second": 4.479, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 4.533965244865719e-05, | |
| "loss": 1.0259, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 4.454976303317536e-05, | |
| "loss": 0.9659, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 4.3759873617693526e-05, | |
| "loss": 0.9996, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.2969984202211694e-05, | |
| "loss": 0.9969, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6259323006310958, | |
| "eval_f1": 0.6150016278628547, | |
| "eval_loss": 0.989003598690033, | |
| "eval_precision": 0.6294369746617599, | |
| "eval_recall": 0.6259323006310958, | |
| "eval_runtime": 24.5638, | |
| "eval_samples_per_second": 70.958, | |
| "eval_steps_per_second": 4.437, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.31, | |
| "learning_rate": 4.218009478672986e-05, | |
| "loss": 0.9251, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.62, | |
| "learning_rate": 4.1390205371248025e-05, | |
| "loss": 0.9081, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "learning_rate": 4.060031595576619e-05, | |
| "loss": 0.9376, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6190476190476191, | |
| "eval_f1": 0.6182633788848451, | |
| "eval_loss": 0.9767513275146484, | |
| "eval_precision": 0.6334597304884556, | |
| "eval_recall": 0.6190476190476191, | |
| "eval_runtime": 24.3243, | |
| "eval_samples_per_second": 71.657, | |
| "eval_steps_per_second": 4.481, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 3.981042654028436e-05, | |
| "loss": 0.8888, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 3.902053712480253e-05, | |
| "loss": 0.891, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "learning_rate": 3.82306477093207e-05, | |
| "loss": 0.8299, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6356855995410212, | |
| "eval_f1": 0.6282041804399863, | |
| "eval_loss": 0.9579089283943176, | |
| "eval_precision": 0.6339366993746611, | |
| "eval_recall": 0.6356855995410212, | |
| "eval_runtime": 24.4946, | |
| "eval_samples_per_second": 71.159, | |
| "eval_steps_per_second": 4.45, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 7.19, | |
| "learning_rate": 3.744075829383886e-05, | |
| "loss": 0.8563, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "learning_rate": 3.665086887835703e-05, | |
| "loss": 0.8477, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 7.81, | |
| "learning_rate": 3.58609794628752e-05, | |
| "loss": 0.7645, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.648881239242685, | |
| "eval_f1": 0.647358676044992, | |
| "eval_loss": 0.9366128444671631, | |
| "eval_precision": 0.6559082514099844, | |
| "eval_recall": 0.648881239242685, | |
| "eval_runtime": 24.292, | |
| "eval_samples_per_second": 71.752, | |
| "eval_steps_per_second": 4.487, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 3.507109004739337e-05, | |
| "loss": 0.8149, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 8.44, | |
| "learning_rate": 3.4281200631911535e-05, | |
| "loss": 0.7755, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "learning_rate": 3.34913112164297e-05, | |
| "loss": 0.7944, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6442914515203672, | |
| "eval_f1": 0.6447425675895946, | |
| "eval_loss": 0.9302862286567688, | |
| "eval_precision": 0.6494435537086374, | |
| "eval_recall": 0.6442914515203672, | |
| "eval_runtime": 24.5079, | |
| "eval_samples_per_second": 71.12, | |
| "eval_steps_per_second": 4.448, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "learning_rate": 3.2701421800947866e-05, | |
| "loss": 0.74, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "learning_rate": 3.1911532385466034e-05, | |
| "loss": 0.7196, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 9.69, | |
| "learning_rate": 3.11216429699842e-05, | |
| "loss": 0.7218, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 3.0331753554502375e-05, | |
| "loss": 0.7334, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6546184738955824, | |
| "eval_f1": 0.6523112247053579, | |
| "eval_loss": 0.9510300755500793, | |
| "eval_precision": 0.6634128950611891, | |
| "eval_recall": 0.6546184738955824, | |
| "eval_runtime": 24.2151, | |
| "eval_samples_per_second": 71.98, | |
| "eval_steps_per_second": 4.501, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 10.31, | |
| "learning_rate": 2.9541864139020537e-05, | |
| "loss": 0.7152, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 10.62, | |
| "learning_rate": 2.8751974723538705e-05, | |
| "loss": 0.6944, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 10.94, | |
| "learning_rate": 2.7962085308056874e-05, | |
| "loss": 0.6596, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.6448651749856569, | |
| "eval_f1": 0.6428493928535022, | |
| "eval_loss": 0.936934232711792, | |
| "eval_precision": 0.6528079277533798, | |
| "eval_recall": 0.6448651749856569, | |
| "eval_runtime": 24.2671, | |
| "eval_samples_per_second": 71.826, | |
| "eval_steps_per_second": 4.492, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 11.25, | |
| "learning_rate": 2.7172195892575043e-05, | |
| "loss": 0.6759, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 11.56, | |
| "learning_rate": 2.638230647709321e-05, | |
| "loss": 0.6274, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 11.88, | |
| "learning_rate": 2.5592417061611373e-05, | |
| "loss": 0.6781, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6368330464716007, | |
| "eval_f1": 0.6359635726412315, | |
| "eval_loss": 0.9716911911964417, | |
| "eval_precision": 0.6513038081890579, | |
| "eval_recall": 0.6368330464716007, | |
| "eval_runtime": 24.1568, | |
| "eval_samples_per_second": 72.154, | |
| "eval_steps_per_second": 4.512, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 12.19, | |
| "learning_rate": 2.480252764612954e-05, | |
| "loss": 0.6889, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "learning_rate": 2.401263823064771e-05, | |
| "loss": 0.6189, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 12.81, | |
| "learning_rate": 2.322274881516588e-05, | |
| "loss": 0.5688, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.6540447504302926, | |
| "eval_f1": 0.6495260815424722, | |
| "eval_loss": 0.95087730884552, | |
| "eval_precision": 0.6531237727480816, | |
| "eval_recall": 0.6540447504302926, | |
| "eval_runtime": 24.3359, | |
| "eval_samples_per_second": 71.623, | |
| "eval_steps_per_second": 4.479, | |
| "step": 416 | |
| }, | |
| { | |
| "epoch": 13.12, | |
| "learning_rate": 2.2432859399684044e-05, | |
| "loss": 0.6399, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 13.44, | |
| "learning_rate": 2.1642969984202213e-05, | |
| "loss": 0.5852, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 13.75, | |
| "learning_rate": 2.0853080568720378e-05, | |
| "loss": 0.5766, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.6615031554790591, | |
| "eval_f1": 0.6601098289619065, | |
| "eval_loss": 0.9484531283378601, | |
| "eval_precision": 0.6655227009747305, | |
| "eval_recall": 0.6615031554790591, | |
| "eval_runtime": 24.0407, | |
| "eval_samples_per_second": 72.502, | |
| "eval_steps_per_second": 4.534, | |
| "step": 448 | |
| }, | |
| { | |
| "epoch": 14.06, | |
| "learning_rate": 2.0063191153238547e-05, | |
| "loss": 0.5559, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 14.38, | |
| "learning_rate": 1.9273301737756715e-05, | |
| "loss": 0.5602, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 14.69, | |
| "learning_rate": 1.848341232227488e-05, | |
| "loss": 0.5529, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 1.769352290679305e-05, | |
| "loss": 0.5529, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.6569133677567413, | |
| "eval_f1": 0.6537964772465467, | |
| "eval_loss": 0.9590263962745667, | |
| "eval_precision": 0.6561006700640171, | |
| "eval_recall": 0.6569133677567413, | |
| "eval_runtime": 24.2565, | |
| "eval_samples_per_second": 71.857, | |
| "eval_steps_per_second": 4.494, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 15.31, | |
| "learning_rate": 1.6903633491311218e-05, | |
| "loss": 0.5288, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 15.62, | |
| "learning_rate": 1.6113744075829386e-05, | |
| "loss": 0.5355, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 15.94, | |
| "learning_rate": 1.532385466034755e-05, | |
| "loss": 0.4998, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.6511761331038439, | |
| "eval_f1": 0.6487852957891203, | |
| "eval_loss": 0.9676968455314636, | |
| "eval_precision": 0.6513647898316787, | |
| "eval_recall": 0.6511761331038439, | |
| "eval_runtime": 23.9784, | |
| "eval_samples_per_second": 72.69, | |
| "eval_steps_per_second": 4.546, | |
| "step": 512 | |
| }, | |
| { | |
| "epoch": 16.25, | |
| "learning_rate": 1.4533965244865718e-05, | |
| "loss": 0.4864, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 16.56, | |
| "learning_rate": 1.3744075829383887e-05, | |
| "loss": 0.5188, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 16.88, | |
| "learning_rate": 1.2954186413902054e-05, | |
| "loss": 0.4908, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.663798049340218, | |
| "eval_f1": 0.6615693208028428, | |
| "eval_loss": 0.9670336842536926, | |
| "eval_precision": 0.6645442403159969, | |
| "eval_recall": 0.663798049340218, | |
| "eval_runtime": 24.1891, | |
| "eval_samples_per_second": 72.057, | |
| "eval_steps_per_second": 4.506, | |
| "step": 544 | |
| }, | |
| { | |
| "epoch": 17.19, | |
| "learning_rate": 1.216429699842022e-05, | |
| "loss": 0.4697, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 17.5, | |
| "learning_rate": 1.137440758293839e-05, | |
| "loss": 0.4791, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 17.81, | |
| "learning_rate": 1.0584518167456558e-05, | |
| "loss": 0.4682, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.6678141135972461, | |
| "eval_f1": 0.668417951352874, | |
| "eval_loss": 0.9634829163551331, | |
| "eval_precision": 0.6707246062463791, | |
| "eval_recall": 0.6678141135972461, | |
| "eval_runtime": 24.0376, | |
| "eval_samples_per_second": 72.511, | |
| "eval_steps_per_second": 4.535, | |
| "step": 576 | |
| }, | |
| { | |
| "epoch": 18.12, | |
| "learning_rate": 9.794628751974725e-06, | |
| "loss": 0.5067, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 18.44, | |
| "learning_rate": 9.004739336492892e-06, | |
| "loss": 0.4401, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 18.75, | |
| "learning_rate": 8.214849921011059e-06, | |
| "loss": 0.4761, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_f1": 0.6658035749901843, | |
| "eval_loss": 0.968035876750946, | |
| "eval_precision": 0.6673820855481607, | |
| "eval_recall": 0.6666666666666666, | |
| "eval_runtime": 24.4115, | |
| "eval_samples_per_second": 71.401, | |
| "eval_steps_per_second": 4.465, | |
| "step": 608 | |
| }, | |
| { | |
| "epoch": 19.06, | |
| "learning_rate": 7.424960505529226e-06, | |
| "loss": 0.4684, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 19.38, | |
| "learning_rate": 6.635071090047394e-06, | |
| "loss": 0.4466, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 19.69, | |
| "learning_rate": 5.845181674565561e-06, | |
| "loss": 0.4507, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 5.055292259083728e-06, | |
| "loss": 0.4161, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.6712564543889845, | |
| "eval_f1": 0.6700650267836397, | |
| "eval_loss": 0.9700806736946106, | |
| "eval_precision": 0.6719394542069612, | |
| "eval_recall": 0.6712564543889845, | |
| "eval_runtime": 24.0864, | |
| "eval_samples_per_second": 72.364, | |
| "eval_steps_per_second": 4.525, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 20.31, | |
| "learning_rate": 4.265402843601897e-06, | |
| "loss": 0.4724, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 20.62, | |
| "learning_rate": 3.4755134281200636e-06, | |
| "loss": 0.4127, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 20.94, | |
| "learning_rate": 2.685624012638231e-06, | |
| "loss": 0.4295, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.6718301778542742, | |
| "eval_f1": 0.6702822325183047, | |
| "eval_loss": 0.971169114112854, | |
| "eval_precision": 0.6698287004001044, | |
| "eval_recall": 0.6718301778542742, | |
| "eval_runtime": 24.5456, | |
| "eval_samples_per_second": 71.011, | |
| "eval_steps_per_second": 4.441, | |
| "step": 672 | |
| }, | |
| { | |
| "epoch": 21.25, | |
| "learning_rate": 1.8957345971563984e-06, | |
| "loss": 0.4618, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 21.56, | |
| "learning_rate": 1.1058451816745656e-06, | |
| "loss": 0.434, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 21.88, | |
| "learning_rate": 3.15955766192733e-07, | |
| "loss": 0.434, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.6706827309236948, | |
| "eval_f1": 0.6690376760558668, | |
| "eval_loss": 0.9755037426948547, | |
| "eval_precision": 0.6705174221778998, | |
| "eval_recall": 0.6706827309236948, | |
| "eval_runtime": 24.2573, | |
| "eval_samples_per_second": 71.855, | |
| "eval_steps_per_second": 4.493, | |
| "step": 704 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "step": 704, | |
| "total_flos": 6.932186220913164e+18, | |
| "train_loss": 0.7904366488483819, | |
| "train_runtime": 4221.7797, | |
| "train_samples_per_second": 21.188, | |
| "train_steps_per_second": 0.167 | |
| } | |
| ], | |
| "max_steps": 704, | |
| "num_train_epochs": 22, | |
| "total_flos": 6.932186220913164e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |