{ "best_metric": 0.9844559585492227, "best_model_checkpoint": "videomae-base-finetuned-soccer-action-recognitionx4/checkpoint-960", "epoch": 39.02066666666666, "eval_steps": 500, "global_step": 1125, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.02, "eval_accuracy": 0.37823834196891193, "eval_loss": 1.3255189657211304, "eval_runtime": 88.7594, "eval_samples_per_second": 2.174, "eval_steps_per_second": 0.552, "step": 28 }, { "epoch": 1.02, "eval_accuracy": 0.5751295336787565, "eval_loss": 0.9903116822242737, "eval_runtime": 34.7202, "eval_samples_per_second": 5.559, "eval_steps_per_second": 1.411, "step": 56 }, { "epoch": 2.02, "learning_rate": 1.4159292035398232e-05, "loss": 1.1078, "step": 80 }, { "epoch": 2.02, "eval_accuracy": 0.8238341968911918, "eval_loss": 0.46475353837013245, "eval_runtime": 33.6288, "eval_samples_per_second": 5.739, "eval_steps_per_second": 1.457, "step": 84 }, { "epoch": 3.03, "eval_accuracy": 0.8341968911917098, "eval_loss": 0.4074859917163849, "eval_runtime": 33.5102, "eval_samples_per_second": 5.759, "eval_steps_per_second": 1.462, "step": 113 }, { "epoch": 4.02, "eval_accuracy": 0.927461139896373, "eval_loss": 0.23006045818328857, "eval_runtime": 34.0136, "eval_samples_per_second": 5.674, "eval_steps_per_second": 1.441, "step": 141 }, { "epoch": 5.02, "learning_rate": 1.907114624505929e-05, "loss": 0.2591, "step": 160 }, { "epoch": 5.02, "eval_accuracy": 0.8549222797927462, "eval_loss": 0.4089970588684082, "eval_runtime": 33.9545, "eval_samples_per_second": 5.684, "eval_steps_per_second": 1.443, "step": 169 }, { "epoch": 6.02, "eval_accuracy": 0.9481865284974094, "eval_loss": 0.1527147889137268, "eval_runtime": 34.2304, "eval_samples_per_second": 5.638, "eval_steps_per_second": 1.431, "step": 197 }, { "epoch": 7.03, "eval_accuracy": 0.9585492227979274, "eval_loss": 0.14126576483249664, "eval_runtime": 34.1805, "eval_samples_per_second": 5.646, "eval_steps_per_second": 1.434, "step": 226 }, { "epoch": 8.01, "learning_rate": 1.74901185770751e-05, "loss": 0.111, "step": 240 }, { "epoch": 8.02, "eval_accuracy": 0.9637305699481865, "eval_loss": 0.1385740041732788, "eval_runtime": 33.69, "eval_samples_per_second": 5.729, "eval_steps_per_second": 1.454, "step": 254 }, { "epoch": 9.02, "eval_accuracy": 0.9430051813471503, "eval_loss": 0.16739489138126373, "eval_runtime": 34.3119, "eval_samples_per_second": 5.625, "eval_steps_per_second": 1.428, "step": 282 }, { "epoch": 10.02, "eval_accuracy": 0.9740932642487047, "eval_loss": 0.10096671432256699, "eval_runtime": 33.6401, "eval_samples_per_second": 5.737, "eval_steps_per_second": 1.457, "step": 310 }, { "epoch": 11.01, "learning_rate": 1.590909090909091e-05, "loss": 0.0509, "step": 320 }, { "epoch": 11.03, "eval_accuracy": 0.9637305699481865, "eval_loss": 0.1586216688156128, "eval_runtime": 33.6589, "eval_samples_per_second": 5.734, "eval_steps_per_second": 1.456, "step": 339 }, { "epoch": 12.02, "eval_accuracy": 0.8860103626943006, "eval_loss": 0.2695624530315399, "eval_runtime": 34.0413, "eval_samples_per_second": 5.67, "eval_steps_per_second": 1.439, "step": 367 }, { "epoch": 13.02, "eval_accuracy": 0.917098445595855, "eval_loss": 0.3004997670650482, "eval_runtime": 33.8891, "eval_samples_per_second": 5.695, "eval_steps_per_second": 1.446, "step": 395 }, { "epoch": 14.0, "learning_rate": 1.432806324110672e-05, "loss": 0.023, "step": 400 }, { "epoch": 14.02, "eval_accuracy": 0.9740932642487047, "eval_loss": 0.10009404271841049, "eval_runtime": 34.3813, "eval_samples_per_second": 5.614, "eval_steps_per_second": 1.425, "step": 423 }, { "epoch": 15.03, "eval_accuracy": 0.9481865284974094, "eval_loss": 0.19607579708099365, "eval_runtime": 33.8549, "eval_samples_per_second": 5.701, "eval_steps_per_second": 1.447, "step": 452 }, { "epoch": 16.02, "learning_rate": 1.274703557312253e-05, "loss": 0.0354, "step": 480 }, { "epoch": 16.02, "eval_accuracy": 0.9222797927461139, "eval_loss": 0.2595587968826294, "eval_runtime": 34.7628, "eval_samples_per_second": 5.552, "eval_steps_per_second": 1.41, "step": 480 }, { "epoch": 17.02, "eval_accuracy": 0.9689119170984456, "eval_loss": 0.10061533749103546, "eval_runtime": 36.7389, "eval_samples_per_second": 5.253, "eval_steps_per_second": 1.334, "step": 508 }, { "epoch": 18.02, "eval_accuracy": 0.9792746113989638, "eval_loss": 0.09470318257808685, "eval_runtime": 35.1648, "eval_samples_per_second": 5.488, "eval_steps_per_second": 1.393, "step": 536 }, { "epoch": 19.02, "learning_rate": 1.116600790513834e-05, "loss": 0.0141, "step": 560 }, { "epoch": 19.03, "eval_accuracy": 0.9792746113989638, "eval_loss": 0.08311115205287933, "eval_runtime": 35.0699, "eval_samples_per_second": 5.503, "eval_steps_per_second": 1.397, "step": 565 }, { "epoch": 20.02, "eval_accuracy": 0.9792746113989638, "eval_loss": 0.06850504875183105, "eval_runtime": 34.7445, "eval_samples_per_second": 5.555, "eval_steps_per_second": 1.41, "step": 593 }, { "epoch": 21.02, "eval_accuracy": 0.9689119170984456, "eval_loss": 0.09778512269258499, "eval_runtime": 34.2526, "eval_samples_per_second": 5.635, "eval_steps_per_second": 1.431, "step": 621 }, { "epoch": 22.02, "learning_rate": 9.58498023715415e-06, "loss": 0.0168, "step": 640 }, { "epoch": 22.02, "eval_accuracy": 0.9792746113989638, "eval_loss": 0.08115241676568985, "eval_runtime": 33.4433, "eval_samples_per_second": 5.771, "eval_steps_per_second": 1.465, "step": 649 }, { "epoch": 23.03, "eval_accuracy": 0.9792746113989638, "eval_loss": 0.07816170156002045, "eval_runtime": 32.8844, "eval_samples_per_second": 5.869, "eval_steps_per_second": 1.49, "step": 678 }, { "epoch": 24.02, "eval_accuracy": 0.927461139896373, "eval_loss": 0.2624199390411377, "eval_runtime": 33.6116, "eval_samples_per_second": 5.742, "eval_steps_per_second": 1.458, "step": 706 }, { "epoch": 25.01, "learning_rate": 8.003952569169962e-06, "loss": 0.0029, "step": 720 }, { "epoch": 25.02, "eval_accuracy": 0.9533678756476683, "eval_loss": 0.1805860698223114, "eval_runtime": 33.2451, "eval_samples_per_second": 5.805, "eval_steps_per_second": 1.474, "step": 734 }, { "epoch": 26.02, "eval_accuracy": 0.9585492227979274, "eval_loss": 0.21230334043502808, "eval_runtime": 32.7523, "eval_samples_per_second": 5.893, "eval_steps_per_second": 1.496, "step": 762 }, { "epoch": 27.03, "eval_accuracy": 0.9481865284974094, "eval_loss": 0.200972780585289, "eval_runtime": 34.3388, "eval_samples_per_second": 5.62, "eval_steps_per_second": 1.427, "step": 791 }, { "epoch": 28.01, "learning_rate": 6.422924901185771e-06, "loss": 0.0048, "step": 800 }, { "epoch": 28.02, "eval_accuracy": 0.9689119170984456, "eval_loss": 0.12585969269275665, "eval_runtime": 32.5431, "eval_samples_per_second": 5.931, "eval_steps_per_second": 1.506, "step": 819 }, { "epoch": 29.02, "eval_accuracy": 0.9430051813471503, "eval_loss": 0.2145320475101471, "eval_runtime": 33.8812, "eval_samples_per_second": 5.696, "eval_steps_per_second": 1.446, "step": 847 }, { "epoch": 30.02, "eval_accuracy": 0.9222797927461139, "eval_loss": 0.29007139801979065, "eval_runtime": 33.0865, "eval_samples_per_second": 5.833, "eval_steps_per_second": 1.481, "step": 875 }, { "epoch": 31.0, "learning_rate": 4.841897233201581e-06, "loss": 0.0042, "step": 880 }, { "epoch": 31.03, "eval_accuracy": 0.9792746113989638, "eval_loss": 0.0911334827542305, "eval_runtime": 32.213, "eval_samples_per_second": 5.991, "eval_steps_per_second": 1.521, "step": 904 }, { "epoch": 32.02, "eval_accuracy": 0.9740932642487047, "eval_loss": 0.0891527533531189, "eval_runtime": 35.2316, "eval_samples_per_second": 5.478, "eval_steps_per_second": 1.391, "step": 932 }, { "epoch": 33.02, "learning_rate": 3.2608695652173914e-06, "loss": 0.0024, "step": 960 }, { "epoch": 33.02, "eval_accuracy": 0.9844559585492227, "eval_loss": 0.09106723964214325, "eval_runtime": 33.6234, "eval_samples_per_second": 5.74, "eval_steps_per_second": 1.457, "step": 960 }, { "epoch": 34.02, "eval_accuracy": 0.9740932642487047, "eval_loss": 0.09867668896913528, "eval_runtime": 34.9535, "eval_samples_per_second": 5.522, "eval_steps_per_second": 1.402, "step": 988 }, { "epoch": 35.03, "eval_accuracy": 0.9481865284974094, "eval_loss": 0.14462456107139587, "eval_runtime": 34.1647, "eval_samples_per_second": 5.649, "eval_steps_per_second": 1.434, "step": 1017 }, { "epoch": 36.02, "learning_rate": 1.6798418972332018e-06, "loss": 0.0008, "step": 1040 }, { "epoch": 36.02, "eval_accuracy": 0.9481865284974094, "eval_loss": 0.14153124392032623, "eval_runtime": 33.7336, "eval_samples_per_second": 5.721, "eval_steps_per_second": 1.453, "step": 1045 }, { "epoch": 37.02, "eval_accuracy": 0.9481865284974094, "eval_loss": 0.1434861421585083, "eval_runtime": 34.3867, "eval_samples_per_second": 5.613, "eval_steps_per_second": 1.425, "step": 1073 }, { "epoch": 38.02, "eval_accuracy": 0.9481865284974094, "eval_loss": 0.14380092918872833, "eval_runtime": 34.1415, "eval_samples_per_second": 5.653, "eval_steps_per_second": 1.435, "step": 1101 }, { "epoch": 39.02, "learning_rate": 9.881422924901187e-08, "loss": 0.0008, "step": 1120 }, { "epoch": 39.02, "eval_accuracy": 0.9533678756476683, "eval_loss": 0.137101411819458, "eval_runtime": 34.8796, "eval_samples_per_second": 5.533, "eval_steps_per_second": 1.405, "step": 1125 }, { "epoch": 39.02, "step": 1125, "total_flos": 4.471333316936427e+19, "train_loss": 0.1162004730856667, "train_runtime": 11133.0856, "train_samples_per_second": 3.234, "train_steps_per_second": 0.101 }, { "epoch": 39.02, "eval_accuracy": 0.964824120603015, "eval_loss": 0.13270962238311768, "eval_runtime": 98.7852, "eval_samples_per_second": 2.014, "eval_steps_per_second": 0.506, "step": 1125 }, { "epoch": 39.02, "eval_accuracy": 0.964824120603015, "eval_loss": 0.13270962238311768, "eval_runtime": 34.085, "eval_samples_per_second": 5.838, "eval_steps_per_second": 1.467, "step": 1125 } ], "logging_steps": 80, "max_steps": 1125, "num_train_epochs": 9223372036854775807, "save_steps": 500, "total_flos": 4.471333316936427e+19, "trial_name": null, "trial_params": null }