| { | |
| "best_metric": 0.9273835920177383, | |
| "best_model_checkpoint": "/home/hoan/projects/nsfw/models/efficientvit_l1.r224_in1k/run-2787ye3h/checkpoint-1147", | |
| "epoch": 26.988235294117647, | |
| "eval_steps": 500, | |
| "global_step": 1147, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 3.626442337305279e-05, | |
| "loss": 1.293, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.8755543237250555, | |
| "eval_loss": 0.9349600672721863, | |
| "eval_runtime": 6.9886, | |
| "eval_samples_per_second": 516.266, | |
| "eval_steps_per_second": 8.156, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 7.252884674610559e-05, | |
| "loss": 0.9064, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 0.00010879327011915836, | |
| "loss": 0.8316, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.914079822616408, | |
| "eval_loss": 0.8049944639205933, | |
| "eval_runtime": 6.9141, | |
| "eval_samples_per_second": 521.829, | |
| "eval_steps_per_second": 8.244, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 0.00014505769349221117, | |
| "loss": 0.7997, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.9126940133037694, | |
| "eval_loss": 0.8133141398429871, | |
| "eval_runtime": 6.6693, | |
| "eval_samples_per_second": 540.987, | |
| "eval_steps_per_second": 8.547, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 0.00018132211686526395, | |
| "loss": 0.7971, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 0.00018108536618532123, | |
| "loss": 0.7806, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9182372505543237, | |
| "eval_loss": 0.8043554425239563, | |
| "eval_runtime": 6.9887, | |
| "eval_samples_per_second": 516.26, | |
| "eval_steps_per_second": 8.156, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 0.00018037635063854584, | |
| "loss": 0.7788, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 0.00017919877324618395, | |
| "loss": 0.764, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_accuracy": 0.917960088691796, | |
| "eval_loss": 0.7976019382476807, | |
| "eval_runtime": 6.7358, | |
| "eval_samples_per_second": 535.642, | |
| "eval_steps_per_second": 8.462, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "learning_rate": 0.00017755878421766717, | |
| "loss": 0.7658, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9016075388026608, | |
| "eval_loss": 0.8304137587547302, | |
| "eval_runtime": 6.8918, | |
| "eval_samples_per_second": 523.522, | |
| "eval_steps_per_second": 8.271, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 0.00017546494882951695, | |
| "loss": 0.751, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.73, | |
| "learning_rate": 0.00017292820269092213, | |
| "loss": 0.7533, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "eval_accuracy": 0.9049334811529933, | |
| "eval_loss": 0.8278865218162537, | |
| "eval_runtime": 6.9038, | |
| "eval_samples_per_second": 522.607, | |
| "eval_steps_per_second": 8.256, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 7.34, | |
| "learning_rate": 0.00016996179462962718, | |
| "loss": 0.747, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 7.95, | |
| "learning_rate": 0.00016658121749642477, | |
| "loss": 0.7294, | |
| "step": 338 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9054878048780488, | |
| "eval_loss": 0.8298162817955017, | |
| "eval_runtime": 6.8854, | |
| "eval_samples_per_second": 524.008, | |
| "eval_steps_per_second": 8.278, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "learning_rate": 0.0001628041272496441, | |
| "loss": 0.7325, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "eval_accuracy": 0.9174057649667405, | |
| "eval_loss": 0.8046008944511414, | |
| "eval_runtime": 6.8095, | |
| "eval_samples_per_second": 529.85, | |
| "eval_steps_per_second": 8.371, | |
| "step": 382 | |
| }, | |
| { | |
| "epoch": 9.18, | |
| "learning_rate": 0.0001586502507422381, | |
| "loss": 0.737, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 0.0001541412826930757, | |
| "loss": 0.7235, | |
| "step": 416 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.917960088691796, | |
| "eval_loss": 0.8011636734008789, | |
| "eval_runtime": 7.0572, | |
| "eval_samples_per_second": 511.252, | |
| "eval_steps_per_second": 8.077, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "learning_rate": 0.00014930077238053366, | |
| "loss": 0.7212, | |
| "step": 442 | |
| }, | |
| { | |
| "epoch": 10.99, | |
| "eval_accuracy": 0.9154656319290465, | |
| "eval_loss": 0.8091404438018799, | |
| "eval_runtime": 6.6144, | |
| "eval_samples_per_second": 545.478, | |
| "eval_steps_per_second": 8.618, | |
| "step": 467 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 0.00014415400065016071, | |
| "loss": 0.7204, | |
| "step": 468 | |
| }, | |
| { | |
| "epoch": 11.62, | |
| "learning_rate": 0.00013872784787877452, | |
| "loss": 0.7154, | |
| "step": 494 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8747228381374723, | |
| "eval_loss": 0.8664525747299194, | |
| "eval_runtime": 6.7378, | |
| "eval_samples_per_second": 535.484, | |
| "eval_steps_per_second": 8.46, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 12.24, | |
| "learning_rate": 0.00013305065358458295, | |
| "loss": 0.7148, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 12.85, | |
| "learning_rate": 0.00012715206841655379, | |
| "loss": 0.7122, | |
| "step": 546 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "eval_accuracy": 0.9154656319290465, | |
| "eval_loss": 0.8063619136810303, | |
| "eval_runtime": 6.798, | |
| "eval_samples_per_second": 530.746, | |
| "eval_steps_per_second": 8.385, | |
| "step": 552 | |
| }, | |
| { | |
| "epoch": 13.46, | |
| "learning_rate": 0.00012106289929605711, | |
| "loss": 0.7096, | |
| "step": 572 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9162971175166297, | |
| "eval_loss": 0.8059455752372742, | |
| "eval_runtime": 6.7495, | |
| "eval_samples_per_second": 534.558, | |
| "eval_steps_per_second": 8.445, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 14.07, | |
| "learning_rate": 0.00011481494851956916, | |
| "loss": 0.7157, | |
| "step": 598 | |
| }, | |
| { | |
| "epoch": 14.68, | |
| "learning_rate": 0.0001084408476627666, | |
| "loss": 0.7021, | |
| "step": 624 | |
| }, | |
| { | |
| "epoch": 14.99, | |
| "eval_accuracy": 0.9212860310421286, | |
| "eval_loss": 0.8017489314079285, | |
| "eval_runtime": 6.9058, | |
| "eval_samples_per_second": 522.457, | |
| "eval_steps_per_second": 8.254, | |
| "step": 637 | |
| }, | |
| { | |
| "epoch": 15.29, | |
| "learning_rate": 0.0001019738871534904, | |
| "loss": 0.7021, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 15.91, | |
| "learning_rate": 9.544784240368045e-05, | |
| "loss": 0.7002, | |
| "step": 676 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9157427937915743, | |
| "eval_loss": 0.816520631313324, | |
| "eval_runtime": 6.8, | |
| "eval_samples_per_second": 530.585, | |
| "eval_steps_per_second": 8.382, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 16.52, | |
| "learning_rate": 8.889679740835216e-05, | |
| "loss": 0.705, | |
| "step": 702 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "eval_accuracy": 0.9196230598669624, | |
| "eval_loss": 0.8022974133491516, | |
| "eval_runtime": 6.9057, | |
| "eval_samples_per_second": 522.468, | |
| "eval_steps_per_second": 8.254, | |
| "step": 722 | |
| }, | |
| { | |
| "epoch": 17.13, | |
| "learning_rate": 8.235496673291634e-05, | |
| "loss": 0.7031, | |
| "step": 728 | |
| }, | |
| { | |
| "epoch": 17.74, | |
| "learning_rate": 7.585651681856075e-05, | |
| "loss": 0.6987, | |
| "step": 754 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9154656319290465, | |
| "eval_loss": 0.8067501783370972, | |
| "eval_runtime": 6.7889, | |
| "eval_samples_per_second": 531.457, | |
| "eval_steps_per_second": 8.396, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 18.35, | |
| "learning_rate": 6.943538753897311e-05, | |
| "loss": 0.6986, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 18.96, | |
| "learning_rate": 6.312511494037267e-05, | |
| "loss": 0.698, | |
| "step": 806 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "eval_accuracy": 0.9262749445676275, | |
| "eval_loss": 0.7983365058898926, | |
| "eval_runtime": 6.7297, | |
| "eval_samples_per_second": 536.129, | |
| "eval_steps_per_second": 8.47, | |
| "step": 807 | |
| }, | |
| { | |
| "epoch": 19.58, | |
| "learning_rate": 5.695865609063774e-05, | |
| "loss": 0.6899, | |
| "step": 832 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9174057649667405, | |
| "eval_loss": 0.8008666038513184, | |
| "eval_runtime": 6.6376, | |
| "eval_samples_per_second": 543.571, | |
| "eval_steps_per_second": 8.587, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 20.19, | |
| "learning_rate": 5.0968216952300726e-05, | |
| "loss": 0.6935, | |
| "step": 858 | |
| }, | |
| { | |
| "epoch": 20.8, | |
| "learning_rate": 4.518508417838943e-05, | |
| "loss": 0.6922, | |
| "step": 884 | |
| }, | |
| { | |
| "epoch": 20.99, | |
| "eval_accuracy": 0.9221175166297118, | |
| "eval_loss": 0.8021017909049988, | |
| "eval_runtime": 6.9244, | |
| "eval_samples_per_second": 521.053, | |
| "eval_steps_per_second": 8.232, | |
| "step": 892 | |
| }, | |
| { | |
| "epoch": 21.41, | |
| "learning_rate": 3.9639461709605394e-05, | |
| "loss": 0.6895, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.9240576496674058, | |
| "eval_loss": 0.7964810729026794, | |
| "eval_runtime": 6.7682, | |
| "eval_samples_per_second": 533.077, | |
| "eval_steps_per_second": 8.422, | |
| "step": 935 | |
| }, | |
| { | |
| "epoch": 22.02, | |
| "learning_rate": 3.436031302625408e-05, | |
| "loss": 0.6875, | |
| "step": 936 | |
| }, | |
| { | |
| "epoch": 22.64, | |
| "learning_rate": 2.937520987880715e-05, | |
| "loss": 0.6865, | |
| "step": 962 | |
| }, | |
| { | |
| "epoch": 22.99, | |
| "eval_accuracy": 0.9229490022172949, | |
| "eval_loss": 0.7992194294929504, | |
| "eval_runtime": 6.8278, | |
| "eval_samples_per_second": 528.427, | |
| "eval_steps_per_second": 8.348, | |
| "step": 977 | |
| }, | |
| { | |
| "epoch": 23.25, | |
| "learning_rate": 2.4710188287142336e-05, | |
| "loss": 0.6886, | |
| "step": 988 | |
| }, | |
| { | |
| "epoch": 23.86, | |
| "learning_rate": 2.0389612560542293e-05, | |
| "loss": 0.6874, | |
| "step": 1014 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.9243348115299335, | |
| "eval_loss": 0.794614851474762, | |
| "eval_runtime": 7.0369, | |
| "eval_samples_per_second": 512.726, | |
| "eval_steps_per_second": 8.1, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 24.47, | |
| "learning_rate": 1.6436048048644487e-05, | |
| "loss": 0.6936, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 24.99, | |
| "eval_accuracy": 0.924889135254989, | |
| "eval_loss": 0.7939379811286926, | |
| "eval_runtime": 6.7889, | |
| "eval_samples_per_second": 531.455, | |
| "eval_steps_per_second": 8.396, | |
| "step": 1062 | |
| }, | |
| { | |
| "epoch": 25.08, | |
| "learning_rate": 1.2870143287933509e-05, | |
| "loss": 0.6864, | |
| "step": 1066 | |
| }, | |
| { | |
| "epoch": 25.69, | |
| "learning_rate": 9.710522159297248e-06, | |
| "loss": 0.6875, | |
| "step": 1092 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.926829268292683, | |
| "eval_loss": 0.7930501699447632, | |
| "eval_runtime": 6.6519, | |
| "eval_samples_per_second": 542.402, | |
| "eval_steps_per_second": 8.569, | |
| "step": 1105 | |
| }, | |
| { | |
| "epoch": 26.31, | |
| "learning_rate": 6.9736866198824056e-06, | |
| "loss": 0.6869, | |
| "step": 1118 | |
| }, | |
| { | |
| "epoch": 26.92, | |
| "learning_rate": 4.67393051725817e-06, | |
| "loss": 0.6863, | |
| "step": 1144 | |
| }, | |
| { | |
| "epoch": 26.99, | |
| "eval_accuracy": 0.9273835920177383, | |
| "eval_loss": 0.7957404851913452, | |
| "eval_runtime": 6.6539, | |
| "eval_samples_per_second": 542.236, | |
| "eval_steps_per_second": 8.566, | |
| "step": 1147 | |
| } | |
| ], | |
| "logging_steps": 26, | |
| "max_steps": 1260, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "total_flos": 0.0, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": { | |
| "_wandb": {}, | |
| "assignments": {}, | |
| "learning_rate": 0.00018132211686526395, | |
| "metric": "eval/loss", | |
| "warmup_ratio": 0.1030284831764748 | |
| } | |
| } | |