| { | |
| "best_metric": 0.924889135254989, | |
| "best_model_checkpoint": "/home/hoan/projects/nsfw/models/efficientvit_l1.r224_in1k/run-29h0ichm/checkpoint-1190", | |
| "epoch": 28.988235294117647, | |
| "eval_steps": 500, | |
| "global_step": 1232, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.3002801275853926e-05, | |
| "loss": 1.4288, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.8284368070953437, | |
| "eval_loss": 0.9695671796798706, | |
| "eval_runtime": 7.0484, | |
| "eval_samples_per_second": 511.888, | |
| "eval_steps_per_second": 8.087, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 2.6005602551707852e-05, | |
| "loss": 1.0169, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 3.900840382756177e-05, | |
| "loss": 0.8905, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8988359201773836, | |
| "eval_loss": 0.8517076969146729, | |
| "eval_runtime": 6.945, | |
| "eval_samples_per_second": 519.507, | |
| "eval_steps_per_second": 8.207, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 5.2011205103415704e-05, | |
| "loss": 0.8329, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.9093680709534369, | |
| "eval_loss": 0.8142409324645996, | |
| "eval_runtime": 7.0042, | |
| "eval_samples_per_second": 515.123, | |
| "eval_steps_per_second": 8.138, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 6.501400637926963e-05, | |
| "loss": 0.8067, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 7.801680765512355e-05, | |
| "loss": 0.7946, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9162971175166297, | |
| "eval_loss": 0.8081430792808533, | |
| "eval_runtime": 6.8448, | |
| "eval_samples_per_second": 527.116, | |
| "eval_steps_per_second": 8.327, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 9.101960893097748e-05, | |
| "loss": 0.7835, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 0.00010402241020683141, | |
| "loss": 0.7776, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_accuracy": 0.9149113082039911, | |
| "eval_loss": 0.8002059459686279, | |
| "eval_runtime": 6.7746, | |
| "eval_samples_per_second": 532.575, | |
| "eval_steps_per_second": 8.414, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "learning_rate": 0.00011702521148268534, | |
| "loss": 0.7733, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9113082039911308, | |
| "eval_loss": 0.8091906309127808, | |
| "eval_runtime": 6.7326, | |
| "eval_samples_per_second": 535.898, | |
| "eval_steps_per_second": 8.466, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 0.00013002801275853926, | |
| "loss": 0.769, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.73, | |
| "learning_rate": 0.0001430308140343932, | |
| "loss": 0.7606, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "eval_accuracy": 0.9176829268292683, | |
| "eval_loss": 0.8018920421600342, | |
| "eval_runtime": 6.7549, | |
| "eval_samples_per_second": 534.135, | |
| "eval_steps_per_second": 8.438, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 7.34, | |
| "learning_rate": 0.00014277948445408833, | |
| "loss": 0.7627, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 7.95, | |
| "learning_rate": 0.00014202726222924556, | |
| "loss": 0.7504, | |
| "step": 338 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9146341463414634, | |
| "eval_loss": 0.7977239489555359, | |
| "eval_runtime": 6.9911, | |
| "eval_samples_per_second": 516.087, | |
| "eval_steps_per_second": 8.153, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "learning_rate": 0.00014077943449179769, | |
| "loss": 0.7386, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "eval_accuracy": 0.7860310421286031, | |
| "eval_loss": 1.0136197805404663, | |
| "eval_runtime": 6.7992, | |
| "eval_samples_per_second": 530.649, | |
| "eval_steps_per_second": 8.383, | |
| "step": 382 | |
| }, | |
| { | |
| "epoch": 9.18, | |
| "learning_rate": 0.00013904477182796176, | |
| "loss": 0.7485, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 0.00013683546663256446, | |
| "loss": 0.7377, | |
| "step": 416 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9054878048780488, | |
| "eval_loss": 0.8229023814201355, | |
| "eval_runtime": 6.8204, | |
| "eval_samples_per_second": 528.999, | |
| "eval_steps_per_second": 8.357, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "learning_rate": 0.0001341670474125579, | |
| "loss": 0.7304, | |
| "step": 442 | |
| }, | |
| { | |
| "epoch": 10.99, | |
| "eval_accuracy": 0.9176829268292683, | |
| "eval_loss": 0.7997891306877136, | |
| "eval_runtime": 6.9681, | |
| "eval_samples_per_second": 517.789, | |
| "eval_steps_per_second": 8.18, | |
| "step": 467 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 0.00013105826964205915, | |
| "loss": 0.7257, | |
| "step": 468 | |
| }, | |
| { | |
| "epoch": 11.62, | |
| "learning_rate": 0.00012753098393605898, | |
| "loss": 0.7229, | |
| "step": 494 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9157427937915743, | |
| "eval_loss": 0.8081395626068115, | |
| "eval_runtime": 6.8807, | |
| "eval_samples_per_second": 524.363, | |
| "eval_steps_per_second": 8.284, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 12.24, | |
| "learning_rate": 0.000123609982469365, | |
| "loss": 0.7229, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 12.85, | |
| "learning_rate": 0.00011932282472025194, | |
| "loss": 0.7213, | |
| "step": 546 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "eval_accuracy": 0.9187915742793792, | |
| "eval_loss": 0.8033608198165894, | |
| "eval_runtime": 6.492, | |
| "eval_samples_per_second": 555.759, | |
| "eval_steps_per_second": 8.78, | |
| "step": 552 | |
| }, | |
| { | |
| "epoch": 13.46, | |
| "learning_rate": 0.00011469964376361164, | |
| "loss": 0.7199, | |
| "step": 572 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9085365853658537, | |
| "eval_loss": 0.8193542957305908, | |
| "eval_runtime": 6.5648, | |
| "eval_samples_per_second": 549.598, | |
| "eval_steps_per_second": 8.683, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 14.07, | |
| "learning_rate": 0.00010977293447510712, | |
| "loss": 0.72, | |
| "step": 598 | |
| }, | |
| { | |
| "epoch": 14.68, | |
| "learning_rate": 0.00010457732513497633, | |
| "loss": 0.7083, | |
| "step": 624 | |
| }, | |
| { | |
| "epoch": 14.99, | |
| "eval_accuracy": 0.8841463414634146, | |
| "eval_loss": 0.8560231328010559, | |
| "eval_runtime": 7.0946, | |
| "eval_samples_per_second": 508.554, | |
| "eval_steps_per_second": 8.034, | |
| "step": 637 | |
| }, | |
| { | |
| "epoch": 15.29, | |
| "learning_rate": 9.914933403681078e-05, | |
| "loss": 0.7122, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 15.91, | |
| "learning_rate": 9.352711281202806e-05, | |
| "loss": 0.7098, | |
| "step": 676 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9115853658536586, | |
| "eval_loss": 0.8119896650314331, | |
| "eval_runtime": 7.0574, | |
| "eval_samples_per_second": 511.234, | |
| "eval_steps_per_second": 8.077, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 16.52, | |
| "learning_rate": 8.775017827413008e-05, | |
| "loss": 0.7091, | |
| "step": 702 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "eval_accuracy": 0.9187915742793792, | |
| "eval_loss": 0.8150137662887573, | |
| "eval_runtime": 6.6809, | |
| "eval_samples_per_second": 540.044, | |
| "eval_steps_per_second": 8.532, | |
| "step": 722 | |
| }, | |
| { | |
| "epoch": 17.13, | |
| "learning_rate": 8.185913466752837e-05, | |
| "loss": 0.7011, | |
| "step": 728 | |
| }, | |
| { | |
| "epoch": 17.74, | |
| "learning_rate": 7.589538827316175e-05, | |
| "loss": 0.7042, | |
| "step": 754 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9171286031042128, | |
| "eval_loss": 0.8104116916656494, | |
| "eval_runtime": 6.7801, | |
| "eval_samples_per_second": 532.147, | |
| "eval_steps_per_second": 8.407, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 18.35, | |
| "learning_rate": 6.990085637685318e-05, | |
| "loss": 0.7005, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 18.96, | |
| "learning_rate": 6.391767264597476e-05, | |
| "loss": 0.7007, | |
| "step": 806 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "eval_accuracy": 0.8980044345898004, | |
| "eval_loss": 0.8331359028816223, | |
| "eval_runtime": 6.6744, | |
| "eval_samples_per_second": 540.575, | |
| "eval_steps_per_second": 8.54, | |
| "step": 807 | |
| }, | |
| { | |
| "epoch": 19.58, | |
| "learning_rate": 5.798789098523564e-05, | |
| "loss": 0.7058, | |
| "step": 832 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.916019955654102, | |
| "eval_loss": 0.8079680800437927, | |
| "eval_runtime": 6.7387, | |
| "eval_samples_per_second": 535.412, | |
| "eval_steps_per_second": 8.459, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 20.19, | |
| "learning_rate": 5.215318995309626e-05, | |
| "loss": 0.6963, | |
| "step": 858 | |
| }, | |
| { | |
| "epoch": 20.8, | |
| "learning_rate": 4.6454579816372486e-05, | |
| "loss": 0.6957, | |
| "step": 884 | |
| }, | |
| { | |
| "epoch": 20.99, | |
| "eval_accuracy": 0.917960088691796, | |
| "eval_loss": 0.810405969619751, | |
| "eval_runtime": 6.9683, | |
| "eval_samples_per_second": 517.773, | |
| "eval_steps_per_second": 8.18, | |
| "step": 892 | |
| }, | |
| { | |
| "epoch": 21.41, | |
| "learning_rate": 4.093211430205049e-05, | |
| "loss": 0.6931, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.917960088691796, | |
| "eval_loss": 0.808899462223053, | |
| "eval_runtime": 6.8846, | |
| "eval_samples_per_second": 524.07, | |
| "eval_steps_per_second": 8.279, | |
| "step": 935 | |
| }, | |
| { | |
| "epoch": 22.02, | |
| "learning_rate": 3.5624609072316884e-05, | |
| "loss": 0.695, | |
| "step": 936 | |
| }, | |
| { | |
| "epoch": 22.64, | |
| "learning_rate": 3.05693689015551e-05, | |
| "loss": 0.693, | |
| "step": 962 | |
| }, | |
| { | |
| "epoch": 22.99, | |
| "eval_accuracy": 0.9223946784922394, | |
| "eval_loss": 0.8023791909217834, | |
| "eval_runtime": 6.7518, | |
| "eval_samples_per_second": 534.373, | |
| "eval_steps_per_second": 8.442, | |
| "step": 977 | |
| }, | |
| { | |
| "epoch": 23.25, | |
| "learning_rate": 2.5801925472893242e-05, | |
| "loss": 0.6939, | |
| "step": 988 | |
| }, | |
| { | |
| "epoch": 23.86, | |
| "learning_rate": 2.1355787637249674e-05, | |
| "loss": 0.6934, | |
| "step": 1014 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.91990022172949, | |
| "eval_loss": 0.8015068769454956, | |
| "eval_runtime": 6.7805, | |
| "eval_samples_per_second": 532.113, | |
| "eval_steps_per_second": 8.406, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 24.47, | |
| "learning_rate": 1.7262205890225772e-05, | |
| "loss": 0.6911, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 24.99, | |
| "eval_accuracy": 0.9185144124168514, | |
| "eval_loss": 0.8050680756568909, | |
| "eval_runtime": 6.6202, | |
| "eval_samples_per_second": 544.999, | |
| "eval_steps_per_second": 8.61, | |
| "step": 1062 | |
| }, | |
| { | |
| "epoch": 25.08, | |
| "learning_rate": 1.3549952722263349e-05, | |
| "loss": 0.6896, | |
| "step": 1066 | |
| }, | |
| { | |
| "epoch": 25.69, | |
| "learning_rate": 1.0245120385916187e-05, | |
| "loss": 0.6879, | |
| "step": 1092 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.9165742793791575, | |
| "eval_loss": 0.8068881034851074, | |
| "eval_runtime": 6.6458, | |
| "eval_samples_per_second": 542.9, | |
| "eval_steps_per_second": 8.577, | |
| "step": 1105 | |
| }, | |
| { | |
| "epoch": 26.31, | |
| "learning_rate": 7.370937501665227e-06, | |
| "loss": 0.6903, | |
| "step": 1118 | |
| }, | |
| { | |
| "epoch": 26.92, | |
| "learning_rate": 4.9476057912973655e-06, | |
| "loss": 0.6932, | |
| "step": 1144 | |
| }, | |
| { | |
| "epoch": 26.99, | |
| "eval_accuracy": 0.9218403547671841, | |
| "eval_loss": 0.8008785247802734, | |
| "eval_runtime": 6.742, | |
| "eval_samples_per_second": 535.156, | |
| "eval_steps_per_second": 8.455, | |
| "step": 1147 | |
| }, | |
| { | |
| "epoch": 27.53, | |
| "learning_rate": 2.9921580863969824e-06, | |
| "loss": 0.6912, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.924889135254989, | |
| "eval_loss": 0.7968100309371948, | |
| "eval_runtime": 6.7991, | |
| "eval_samples_per_second": 530.661, | |
| "eval_steps_per_second": 8.383, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 28.14, | |
| "learning_rate": 1.5183386099637947e-06, | |
| "loss": 0.6861, | |
| "step": 1196 | |
| }, | |
| { | |
| "epoch": 28.75, | |
| "learning_rate": 5.365063726196294e-07, | |
| "loss": 0.6877, | |
| "step": 1222 | |
| }, | |
| { | |
| "epoch": 28.99, | |
| "eval_accuracy": 0.917960088691796, | |
| "eval_loss": 0.8051350712776184, | |
| "eval_runtime": 6.5909, | |
| "eval_samples_per_second": 547.421, | |
| "eval_steps_per_second": 8.648, | |
| "step": 1232 | |
| } | |
| ], | |
| "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": 0.0001430308140343932, | |
| "metric": "eval/loss", | |
| "warmup_ratio": 0.22639169356076547 | |
| } | |
| } | |