| { | |
| "best_metric": 0.9246119733924612, | |
| "best_model_checkpoint": "/home/hoan/projects/nsfw/models/efficientvit_l1.r224_in1k/run-3ekxh3sa/checkpoint-1260", | |
| "epoch": 29.647058823529413, | |
| "eval_steps": 500, | |
| "global_step": 1260, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.721582447456803e-05, | |
| "loss": 1.3991, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.8295454545454546, | |
| "eval_loss": 0.9406095743179321, | |
| "eval_runtime": 6.7839, | |
| "eval_samples_per_second": 531.849, | |
| "eval_steps_per_second": 8.402, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 3.443164894913606e-05, | |
| "loss": 0.975, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 5.16474734237041e-05, | |
| "loss": 0.8697, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9049334811529933, | |
| "eval_loss": 0.8338781595230103, | |
| "eval_runtime": 6.896, | |
| "eval_samples_per_second": 523.2, | |
| "eval_steps_per_second": 8.266, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 6.886329789827213e-05, | |
| "loss": 0.8206, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.9121396895787139, | |
| "eval_loss": 0.8089684844017029, | |
| "eval_runtime": 6.8879, | |
| "eval_samples_per_second": 523.816, | |
| "eval_steps_per_second": 8.275, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 8.607912237284016e-05, | |
| "loss": 0.8023, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 9.931490397010198e-05, | |
| "loss": 0.7878, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9124168514412417, | |
| "eval_loss": 0.8339745402336121, | |
| "eval_runtime": 6.9242, | |
| "eval_samples_per_second": 521.075, | |
| "eval_steps_per_second": 8.232, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 9.91185276336997e-05, | |
| "loss": 0.7811, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 9.865445854143955e-05, | |
| "loss": 0.7735, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_accuracy": 0.916019955654102, | |
| "eval_loss": 0.8007537722587585, | |
| "eval_runtime": 6.7698, | |
| "eval_samples_per_second": 532.955, | |
| "eval_steps_per_second": 8.42, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "learning_rate": 9.792520850539316e-05, | |
| "loss": 0.7664, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9132483370288248, | |
| "eval_loss": 0.8072662949562073, | |
| "eval_runtime": 6.9314, | |
| "eval_samples_per_second": 520.533, | |
| "eval_steps_per_second": 8.224, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 9.69347246512672e-05, | |
| "loss": 0.7558, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.73, | |
| "learning_rate": 9.568836805426145e-05, | |
| "loss": 0.7517, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "eval_accuracy": 0.916019955654102, | |
| "eval_loss": 0.807176411151886, | |
| "eval_runtime": 6.6089, | |
| "eval_samples_per_second": 545.929, | |
| "eval_steps_per_second": 8.625, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 7.34, | |
| "learning_rate": 9.419288472180897e-05, | |
| "loss": 0.7486, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 7.95, | |
| "learning_rate": 9.245636908025691e-05, | |
| "loss": 0.742, | |
| "step": 338 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9143569844789357, | |
| "eval_loss": 0.8011844158172607, | |
| "eval_runtime": 6.62, | |
| "eval_samples_per_second": 545.017, | |
| "eval_steps_per_second": 8.61, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "learning_rate": 9.048822016311923e-05, | |
| "loss": 0.7303, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "eval_accuracy": 0.9099223946784922, | |
| "eval_loss": 0.8109549283981323, | |
| "eval_runtime": 6.4309, | |
| "eval_samples_per_second": 561.046, | |
| "eval_steps_per_second": 8.864, | |
| "step": 382 | |
| }, | |
| { | |
| "epoch": 9.18, | |
| "learning_rate": 8.829909073803584e-05, | |
| "loss": 0.7404, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 8.590082964779303e-05, | |
| "loss": 0.7277, | |
| "step": 416 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9196230598669624, | |
| "eval_loss": 0.7979661822319031, | |
| "eval_runtime": 7.0683, | |
| "eval_samples_per_second": 510.446, | |
| "eval_steps_per_second": 8.064, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "learning_rate": 8.330641767748885e-05, | |
| "loss": 0.724, | |
| "step": 442 | |
| }, | |
| { | |
| "epoch": 10.99, | |
| "eval_accuracy": 0.9165742793791575, | |
| "eval_loss": 0.7978601455688477, | |
| "eval_runtime": 6.6977, | |
| "eval_samples_per_second": 538.694, | |
| "eval_steps_per_second": 8.51, | |
| "step": 467 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 8.05298972949677e-05, | |
| "loss": 0.7215, | |
| "step": 468 | |
| }, | |
| { | |
| "epoch": 11.62, | |
| "learning_rate": 7.758629664480999e-05, | |
| "loss": 0.7184, | |
| "step": 494 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9204545454545454, | |
| "eval_loss": 0.7982057929039001, | |
| "eval_runtime": 6.7476, | |
| "eval_samples_per_second": 534.71, | |
| "eval_steps_per_second": 8.447, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 12.24, | |
| "learning_rate": 7.449154820726551e-05, | |
| "loss": 0.7184, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 12.85, | |
| "learning_rate": 7.126240256239565e-05, | |
| "loss": 0.7189, | |
| "step": 546 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "eval_accuracy": 0.91990022172949, | |
| "eval_loss": 0.7983158826828003, | |
| "eval_runtime": 6.7226, | |
| "eval_samples_per_second": 536.698, | |
| "eval_steps_per_second": 8.479, | |
| "step": 552 | |
| }, | |
| { | |
| "epoch": 13.46, | |
| "learning_rate": 6.791633772618323e-05, | |
| "loss": 0.7161, | |
| "step": 572 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9107538802660754, | |
| "eval_loss": 0.8047741055488586, | |
| "eval_runtime": 6.9456, | |
| "eval_samples_per_second": 519.466, | |
| "eval_steps_per_second": 8.207, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 14.07, | |
| "learning_rate": 6.447146454933571e-05, | |
| "loss": 0.7174, | |
| "step": 598 | |
| }, | |
| { | |
| "epoch": 14.68, | |
| "learning_rate": 6.094642869081846e-05, | |
| "loss": 0.7072, | |
| "step": 624 | |
| }, | |
| { | |
| "epoch": 14.99, | |
| "eval_accuracy": 0.9154656319290465, | |
| "eval_loss": 0.8094056248664856, | |
| "eval_runtime": 6.7463, | |
| "eval_samples_per_second": 534.814, | |
| "eval_steps_per_second": 8.449, | |
| "step": 637 | |
| }, | |
| { | |
| "epoch": 15.29, | |
| "learning_rate": 5.736030969669442e-05, | |
| "loss": 0.7104, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 15.91, | |
| "learning_rate": 5.3732517730514215e-05, | |
| "loss": 0.7057, | |
| "step": 676 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9110310421286031, | |
| "eval_loss": 0.8126114010810852, | |
| "eval_runtime": 6.8439, | |
| "eval_samples_per_second": 527.186, | |
| "eval_steps_per_second": 8.329, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 16.52, | |
| "learning_rate": 5.008268851421187e-05, | |
| "loss": 0.7075, | |
| "step": 702 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "eval_accuracy": 0.9171286031042128, | |
| "eval_loss": 0.8077898621559143, | |
| "eval_runtime": 6.6053, | |
| "eval_samples_per_second": 546.231, | |
| "eval_steps_per_second": 8.629, | |
| "step": 722 | |
| }, | |
| { | |
| "epoch": 17.13, | |
| "learning_rate": 4.64305770481475e-05, | |
| "loss": 0.7026, | |
| "step": 728 | |
| }, | |
| { | |
| "epoch": 17.74, | |
| "learning_rate": 4.279595068554542e-05, | |
| "loss": 0.7021, | |
| "step": 754 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9212860310421286, | |
| "eval_loss": 0.8008798360824585, | |
| "eval_runtime": 6.9773, | |
| "eval_samples_per_second": 517.108, | |
| "eval_steps_per_second": 8.169, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 18.35, | |
| "learning_rate": 3.919848214007159e-05, | |
| "loss": 0.7, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 18.96, | |
| "learning_rate": 3.565764300565529e-05, | |
| "loss": 0.7027, | |
| "step": 806 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "eval_accuracy": 0.9162971175166297, | |
| "eval_loss": 0.8068927526473999, | |
| "eval_runtime": 6.8547, | |
| "eval_samples_per_second": 526.354, | |
| "eval_steps_per_second": 8.315, | |
| "step": 807 | |
| }, | |
| { | |
| "epoch": 19.58, | |
| "learning_rate": 3.2192598364888074e-05, | |
| "loss": 0.7046, | |
| "step": 832 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9193458980044346, | |
| "eval_loss": 0.7985545992851257, | |
| "eval_runtime": 6.7345, | |
| "eval_samples_per_second": 535.751, | |
| "eval_steps_per_second": 8.464, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 20.19, | |
| "learning_rate": 2.8822103056440263e-05, | |
| "loss": 0.6975, | |
| "step": 858 | |
| }, | |
| { | |
| "epoch": 20.8, | |
| "learning_rate": 2.5564400162956554e-05, | |
| "loss": 0.6948, | |
| "step": 884 | |
| }, | |
| { | |
| "epoch": 20.99, | |
| "eval_accuracy": 0.9196230598669624, | |
| "eval_loss": 0.8047547340393066, | |
| "eval_runtime": 6.9281, | |
| "eval_samples_per_second": 520.781, | |
| "eval_steps_per_second": 8.227, | |
| "step": 892 | |
| }, | |
| { | |
| "epoch": 21.41, | |
| "learning_rate": 2.243712226887305e-05, | |
| "loss": 0.6955, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.9168514412416852, | |
| "eval_loss": 0.8027793765068054, | |
| "eval_runtime": 6.7629, | |
| "eval_samples_per_second": 533.497, | |
| "eval_steps_per_second": 8.428, | |
| "step": 935 | |
| }, | |
| { | |
| "epoch": 22.02, | |
| "learning_rate": 1.9457196022605978e-05, | |
| "loss": 0.6981, | |
| "step": 936 | |
| }, | |
| { | |
| "epoch": 22.64, | |
| "learning_rate": 1.6640750519677137e-05, | |
| "loss": 0.6955, | |
| "step": 962 | |
| }, | |
| { | |
| "epoch": 22.99, | |
| "eval_accuracy": 0.9223946784922394, | |
| "eval_loss": 0.800028383731842, | |
| "eval_runtime": 6.8203, | |
| "eval_samples_per_second": 529.009, | |
| "eval_steps_per_second": 8.357, | |
| "step": 977 | |
| }, | |
| { | |
| "epoch": 23.25, | |
| "learning_rate": 1.400303000265985e-05, | |
| "loss": 0.6957, | |
| "step": 988 | |
| }, | |
| { | |
| "epoch": 23.86, | |
| "learning_rate": 1.1558311350464133e-05, | |
| "loss": 0.6954, | |
| "step": 1014 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.9207317073170732, | |
| "eval_loss": 0.797659158706665, | |
| "eval_runtime": 6.7863, | |
| "eval_samples_per_second": 531.656, | |
| "eval_steps_per_second": 8.399, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 24.47, | |
| "learning_rate": 9.319826803557244e-06, | |
| "loss": 0.6908, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 24.99, | |
| "eval_accuracy": 0.9201773835920177, | |
| "eval_loss": 0.8041421175003052, | |
| "eval_runtime": 6.6775, | |
| "eval_samples_per_second": 540.324, | |
| "eval_steps_per_second": 8.536, | |
| "step": 1062 | |
| }, | |
| { | |
| "epoch": 25.08, | |
| "learning_rate": 7.299692343375395e-06, | |
| "loss": 0.6909, | |
| "step": 1066 | |
| }, | |
| { | |
| "epoch": 25.69, | |
| "learning_rate": 5.508842113579347e-06, | |
| "loss": 0.6899, | |
| "step": 1092 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.9149113082039911, | |
| "eval_loss": 0.8050050139427185, | |
| "eval_runtime": 6.6773, | |
| "eval_samples_per_second": 540.337, | |
| "eval_steps_per_second": 8.536, | |
| "step": 1105 | |
| }, | |
| { | |
| "epoch": 26.31, | |
| "learning_rate": 3.956969238103917e-06, | |
| "loss": 0.6924, | |
| "step": 1118 | |
| }, | |
| { | |
| "epoch": 26.92, | |
| "learning_rate": 2.6524733563286136e-06, | |
| "loss": 0.6957, | |
| "step": 1144 | |
| }, | |
| { | |
| "epoch": 26.99, | |
| "eval_accuracy": 0.9243348115299335, | |
| "eval_loss": 0.797722578048706, | |
| "eval_runtime": 6.823, | |
| "eval_samples_per_second": 528.799, | |
| "eval_steps_per_second": 8.354, | |
| "step": 1147 | |
| }, | |
| { | |
| "epoch": 27.53, | |
| "learning_rate": 1.6024151593397756e-06, | |
| "loss": 0.6922, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.9232261640798226, | |
| "eval_loss": 0.794511079788208, | |
| "eval_runtime": 6.8758, | |
| "eval_samples_per_second": 524.736, | |
| "eval_steps_per_second": 8.29, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 28.14, | |
| "learning_rate": 8.124781733602129e-07, | |
| "loss": 0.6893, | |
| "step": 1196 | |
| }, | |
| { | |
| "epoch": 28.75, | |
| "learning_rate": 2.869379971967787e-07, | |
| "loss": 0.6899, | |
| "step": 1222 | |
| }, | |
| { | |
| "epoch": 28.99, | |
| "eval_accuracy": 0.9171286031042128, | |
| "eval_loss": 0.8038355708122253, | |
| "eval_runtime": 6.8749, | |
| "eval_samples_per_second": 524.81, | |
| "eval_steps_per_second": 8.291, | |
| "step": 1232 | |
| }, | |
| { | |
| "epoch": 29.36, | |
| "learning_rate": 2.8639160210733933e-08, | |
| "loss": 0.693, | |
| "step": 1248 | |
| }, | |
| { | |
| "epoch": 29.65, | |
| "eval_accuracy": 0.9246119733924612, | |
| "eval_loss": 0.7983114123344421, | |
| "eval_runtime": 7.0129, | |
| "eval_samples_per_second": 514.48, | |
| "eval_steps_per_second": 8.128, | |
| "step": 1260 | |
| } | |
| ], | |
| "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": 9.932206427635402e-05, | |
| "metric": "eval/loss", | |
| "warmup_ratio": 0.11897323802421422 | |
| } | |
| } | |