| { | |
| "best_metric": 0.24527826743893225, | |
| "best_model_checkpoint": "videomae-base-finetuned-ucfcrime-full2/checkpoint-400", | |
| "epoch": 7.125, | |
| "eval_steps": 500, | |
| "global_step": 800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0125, | |
| "grad_norm": 416496.125, | |
| "learning_rate": 6.25e-06, | |
| "loss": 1.9739, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.025, | |
| "grad_norm": 594418.0625, | |
| "learning_rate": 1.25e-05, | |
| "loss": 2.0117, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0375, | |
| "grad_norm": 669342.5, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 1.9919, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 686419.375, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.8385, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0625, | |
| "grad_norm": 561864.3125, | |
| "learning_rate": 3.125e-05, | |
| "loss": 2.0993, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.075, | |
| "grad_norm": 931406.9375, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 2.1469, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0875, | |
| "grad_norm": 518422.0, | |
| "learning_rate": 4.375e-05, | |
| "loss": 2.2427, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 702745.9375, | |
| "learning_rate": 5e-05, | |
| "loss": 2.0093, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.1125, | |
| "grad_norm": 737373.75, | |
| "learning_rate": 4.930555555555556e-05, | |
| "loss": 2.0531, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 951342.4375, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 2.2039, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "eval_accuracy": 0.16441441441441443, | |
| "eval_loss": 2.4596593379974365, | |
| "eval_runtime": 1486.363, | |
| "eval_samples_per_second": 5.377, | |
| "eval_steps_per_second": 0.672, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0125, | |
| "grad_norm": 783028.0625, | |
| "learning_rate": 4.791666666666667e-05, | |
| "loss": 1.9407, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.025, | |
| "grad_norm": 592436.375, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 2.128, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0375, | |
| "grad_norm": 551391.1875, | |
| "learning_rate": 4.652777777777778e-05, | |
| "loss": 1.8403, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 732315.75, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 2.199, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.0625, | |
| "grad_norm": 833000.5, | |
| "learning_rate": 4.5138888888888894e-05, | |
| "loss": 2.1457, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.075, | |
| "grad_norm": 645952.25, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 2.1546, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.0875, | |
| "grad_norm": 741493.375, | |
| "learning_rate": 4.375e-05, | |
| "loss": 1.9128, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 610463.875, | |
| "learning_rate": 4.305555555555556e-05, | |
| "loss": 1.9581, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.1125, | |
| "grad_norm": 669701.875, | |
| "learning_rate": 4.236111111111111e-05, | |
| "loss": 2.2864, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.125, | |
| "grad_norm": 555390.8125, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.8182, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.125, | |
| "eval_accuracy": 0.2046084109796021, | |
| "eval_loss": 2.3359453678131104, | |
| "eval_runtime": 1481.2653, | |
| "eval_samples_per_second": 5.362, | |
| "eval_steps_per_second": 0.67, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0125, | |
| "grad_norm": 982366.0, | |
| "learning_rate": 4.0972222222222225e-05, | |
| "loss": 1.7361, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.025, | |
| "grad_norm": 489943.53125, | |
| "learning_rate": 4.027777777777778e-05, | |
| "loss": 1.9965, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.0375, | |
| "grad_norm": 691774.0625, | |
| "learning_rate": 3.958333333333333e-05, | |
| "loss": 1.9369, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 592585.875, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 2.2331, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.0625, | |
| "grad_norm": 774321.4375, | |
| "learning_rate": 3.8194444444444444e-05, | |
| "loss": 1.9108, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.075, | |
| "grad_norm": 443132.03125, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 1.5835, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.0875, | |
| "grad_norm": 816018.125, | |
| "learning_rate": 3.6805555555555556e-05, | |
| "loss": 1.9894, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 558110.125, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 1.8828, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.1125, | |
| "grad_norm": 582578.75, | |
| "learning_rate": 3.541666666666667e-05, | |
| "loss": 1.9, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.125, | |
| "grad_norm": 637794.125, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 1.9124, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.125, | |
| "eval_accuracy": 0.2077562326869806, | |
| "eval_loss": 2.680464506149292, | |
| "eval_runtime": 1473.8252, | |
| "eval_samples_per_second": 5.389, | |
| "eval_steps_per_second": 0.674, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0125, | |
| "grad_norm": 860386.8125, | |
| "learning_rate": 3.402777777777778e-05, | |
| "loss": 1.7279, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.025, | |
| "grad_norm": 607838.0, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.7094, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.0375, | |
| "grad_norm": 742987.3125, | |
| "learning_rate": 3.263888888888889e-05, | |
| "loss": 1.9674, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "grad_norm": 1233058.5, | |
| "learning_rate": 3.194444444444444e-05, | |
| "loss": 1.7507, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.0625, | |
| "grad_norm": 582323.3125, | |
| "learning_rate": 3.125e-05, | |
| "loss": 1.7098, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.075, | |
| "grad_norm": 1024568.625, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 1.9264, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.0875, | |
| "grad_norm": 656340.6875, | |
| "learning_rate": 2.9861111111111113e-05, | |
| "loss": 1.6687, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 1037552.0, | |
| "learning_rate": 2.916666666666667e-05, | |
| "loss": 1.6596, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.1125, | |
| "grad_norm": 680651.3125, | |
| "learning_rate": 2.8472222222222223e-05, | |
| "loss": 1.8054, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "grad_norm": 687134.875, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 1.7017, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "eval_accuracy": 0.24527826743893225, | |
| "eval_loss": 2.333141803741455, | |
| "eval_runtime": 1475.5703, | |
| "eval_samples_per_second": 5.382, | |
| "eval_steps_per_second": 0.673, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.0125, | |
| "grad_norm": 719175.6875, | |
| "learning_rate": 2.7083333333333332e-05, | |
| "loss": 1.5792, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 4.025, | |
| "grad_norm": 613320.625, | |
| "learning_rate": 2.6388888888888892e-05, | |
| "loss": 1.7086, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 4.0375, | |
| "grad_norm": 795721.5625, | |
| "learning_rate": 2.5694444444444445e-05, | |
| "loss": 1.7548, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "grad_norm": 754591.0, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.7227, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 4.0625, | |
| "grad_norm": 898280.9375, | |
| "learning_rate": 2.4305555555555558e-05, | |
| "loss": 1.7121, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.075, | |
| "grad_norm": 973403.75, | |
| "learning_rate": 2.361111111111111e-05, | |
| "loss": 1.6087, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.0875, | |
| "grad_norm": 794706.0625, | |
| "learning_rate": 2.2916666666666667e-05, | |
| "loss": 1.6259, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "grad_norm": 780511.9375, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 1.7734, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 4.1125, | |
| "grad_norm": 1010192.125, | |
| "learning_rate": 2.152777777777778e-05, | |
| "loss": 1.6318, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 4.125, | |
| "grad_norm": 610750.1875, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 1.68, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.125, | |
| "eval_accuracy": 0.22248803827751196, | |
| "eval_loss": 2.2636775970458984, | |
| "eval_runtime": 1479.3494, | |
| "eval_samples_per_second": 5.369, | |
| "eval_steps_per_second": 0.671, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.0125, | |
| "grad_norm": 1084686.25, | |
| "learning_rate": 2.013888888888889e-05, | |
| "loss": 1.4838, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 5.025, | |
| "grad_norm": 723063.5625, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 1.7148, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 5.0375, | |
| "grad_norm": 780978.0, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 1.6672, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "grad_norm": 875436.0625, | |
| "learning_rate": 1.8055555555555555e-05, | |
| "loss": 1.5467, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 5.0625, | |
| "grad_norm": 742270.875, | |
| "learning_rate": 1.736111111111111e-05, | |
| "loss": 1.5539, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 5.075, | |
| "grad_norm": 846790.625, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 1.5211, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 5.0875, | |
| "grad_norm": 833175.75, | |
| "learning_rate": 1.597222222222222e-05, | |
| "loss": 1.5436, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "grad_norm": 1164196.0, | |
| "learning_rate": 1.527777777777778e-05, | |
| "loss": 1.2434, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 5.1125, | |
| "grad_norm": 1534336.75, | |
| "learning_rate": 1.4583333333333335e-05, | |
| "loss": 1.6604, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 5.125, | |
| "grad_norm": 1043594.25, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 1.5457, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.125, | |
| "eval_accuracy": 0.22840594308738352, | |
| "eval_loss": 2.471050977706909, | |
| "eval_runtime": 1474.9621, | |
| "eval_samples_per_second": 5.385, | |
| "eval_steps_per_second": 0.673, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.0125, | |
| "grad_norm": 970448.25, | |
| "learning_rate": 1.3194444444444446e-05, | |
| "loss": 1.4714, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.025, | |
| "grad_norm": 623090.375, | |
| "learning_rate": 1.25e-05, | |
| "loss": 1.3935, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 6.0375, | |
| "grad_norm": 744650.8125, | |
| "learning_rate": 1.1805555555555555e-05, | |
| "loss": 1.4471, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "grad_norm": 971446.3125, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 1.5173, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 6.0625, | |
| "grad_norm": 773394.0625, | |
| "learning_rate": 1.0416666666666668e-05, | |
| "loss": 1.4809, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 6.075, | |
| "grad_norm": 852698.6875, | |
| "learning_rate": 9.722222222222223e-06, | |
| "loss": 1.392, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 6.0875, | |
| "grad_norm": 949506.75, | |
| "learning_rate": 9.027777777777777e-06, | |
| "loss": 1.4143, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "grad_norm": 1045740.0, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 1.6005, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 6.1125, | |
| "grad_norm": 793035.75, | |
| "learning_rate": 7.63888888888889e-06, | |
| "loss": 1.3571, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 6.125, | |
| "grad_norm": 638157.5625, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 1.4727, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.125, | |
| "eval_accuracy": 0.21102996726265424, | |
| "eval_loss": 2.369114637374878, | |
| "eval_runtime": 1473.3779, | |
| "eval_samples_per_second": 5.39, | |
| "eval_steps_per_second": 0.674, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.0125, | |
| "grad_norm": 529226.1875, | |
| "learning_rate": 6.25e-06, | |
| "loss": 1.0833, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 7.025, | |
| "grad_norm": 796843.1875, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 1.341, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 7.0375, | |
| "grad_norm": 747058.1875, | |
| "learning_rate": 4.861111111111111e-06, | |
| "loss": 1.5642, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "grad_norm": 647754.0625, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 1.3922, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 7.0625, | |
| "grad_norm": 1013934.3125, | |
| "learning_rate": 3.4722222222222224e-06, | |
| "loss": 1.5842, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 7.075, | |
| "grad_norm": 694990.5625, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 1.2497, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 7.0875, | |
| "grad_norm": 964756.0, | |
| "learning_rate": 2.0833333333333334e-06, | |
| "loss": 1.2914, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "grad_norm": 754335.25, | |
| "learning_rate": 1.388888888888889e-06, | |
| "loss": 1.3096, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 7.1125, | |
| "grad_norm": 1201313.375, | |
| "learning_rate": 6.944444444444445e-07, | |
| "loss": 1.4434, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 7.125, | |
| "grad_norm": 1206534.625, | |
| "learning_rate": 0.0, | |
| "loss": 1.5393, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.125, | |
| "eval_accuracy": 0.20246789221858474, | |
| "eval_loss": 2.456111431121826, | |
| "eval_runtime": 1474.6345, | |
| "eval_samples_per_second": 5.386, | |
| "eval_steps_per_second": 0.673, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.125, | |
| "step": 800, | |
| "total_flos": 7.975666818691891e+18, | |
| "train_loss": 1.7285779798030854, | |
| "train_runtime": 14999.7991, | |
| "train_samples_per_second": 0.427, | |
| "train_steps_per_second": 0.053 | |
| }, | |
| { | |
| "epoch": 7.125, | |
| "eval_accuracy": 0.21431451612903227, | |
| "eval_loss": 2.547022819519043, | |
| "eval_runtime": 883.8934, | |
| "eval_samples_per_second": 5.612, | |
| "eval_steps_per_second": 0.701, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.125, | |
| "eval_accuracy": 0.21431451612903227, | |
| "eval_loss": 2.547022581100464, | |
| "eval_runtime": 875.9015, | |
| "eval_samples_per_second": 5.663, | |
| "eval_steps_per_second": 0.708, | |
| "step": 800 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 800, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.975666818691891e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |