{
"best_metric": 0.9428571428571428,
"best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-1200",
"epoch": 3.25,
"global_step": 1200,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 4.166666666666667e-06,
"loss": 2.3416,
"step": 10
},
{
"epoch": 0.02,
"learning_rate": 8.333333333333334e-06,
"loss": 2.3668,
"step": 20
},
{
"epoch": 0.03,
"learning_rate": 1.25e-05,
"loss": 2.4476,
"step": 30
},
{
"epoch": 0.03,
"learning_rate": 1.6666666666666667e-05,
"loss": 2.2052,
"step": 40
},
{
"epoch": 0.04,
"learning_rate": 2.0833333333333336e-05,
"loss": 2.5474,
"step": 50
},
{
"epoch": 0.05,
"learning_rate": 2.5e-05,
"loss": 2.163,
"step": 60
},
{
"epoch": 0.06,
"learning_rate": 2.916666666666667e-05,
"loss": 2.0722,
"step": 70
},
{
"epoch": 0.07,
"learning_rate": 3.3333333333333335e-05,
"loss": 2.3828,
"step": 80
},
{
"epoch": 0.07,
"learning_rate": 3.7500000000000003e-05,
"loss": 1.8969,
"step": 90
},
{
"epoch": 0.08,
"learning_rate": 4.166666666666667e-05,
"loss": 2.2401,
"step": 100
},
{
"epoch": 0.09,
"learning_rate": 4.5833333333333334e-05,
"loss": 2.3168,
"step": 110
},
{
"epoch": 0.1,
"learning_rate": 5e-05,
"loss": 2.1487,
"step": 120
},
{
"epoch": 0.11,
"learning_rate": 4.9537037037037035e-05,
"loss": 2.1718,
"step": 130
},
{
"epoch": 0.12,
"learning_rate": 4.9074074074074075e-05,
"loss": 2.2061,
"step": 140
},
{
"epoch": 0.12,
"learning_rate": 4.8611111111111115e-05,
"loss": 1.829,
"step": 150
},
{
"epoch": 0.13,
"learning_rate": 4.814814814814815e-05,
"loss": 1.8218,
"step": 160
},
{
"epoch": 0.14,
"learning_rate": 4.768518518518519e-05,
"loss": 1.8595,
"step": 170
},
{
"epoch": 0.15,
"learning_rate": 4.722222222222222e-05,
"loss": 1.8735,
"step": 180
},
{
"epoch": 0.16,
"learning_rate": 4.675925925925926e-05,
"loss": 2.1193,
"step": 190
},
{
"epoch": 0.17,
"learning_rate": 4.62962962962963e-05,
"loss": 1.5221,
"step": 200
},
{
"epoch": 0.17,
"learning_rate": 4.5833333333333334e-05,
"loss": 2.0932,
"step": 210
},
{
"epoch": 0.18,
"learning_rate": 4.5370370370370374e-05,
"loss": 1.2333,
"step": 220
},
{
"epoch": 0.19,
"learning_rate": 4.490740740740741e-05,
"loss": 1.7413,
"step": 230
},
{
"epoch": 0.2,
"learning_rate": 4.4444444444444447e-05,
"loss": 2.2578,
"step": 240
},
{
"epoch": 0.21,
"learning_rate": 4.3981481481481486e-05,
"loss": 1.5015,
"step": 250
},
{
"epoch": 0.22,
"learning_rate": 4.351851851851852e-05,
"loss": 1.4546,
"step": 260
},
{
"epoch": 0.23,
"learning_rate": 4.305555555555556e-05,
"loss": 1.5539,
"step": 270
},
{
"epoch": 0.23,
"learning_rate": 4.259259259259259e-05,
"loss": 1.4377,
"step": 280
},
{
"epoch": 0.24,
"learning_rate": 4.212962962962963e-05,
"loss": 0.9734,
"step": 290
},
{
"epoch": 0.25,
"learning_rate": 4.166666666666667e-05,
"loss": 1.3834,
"step": 300
},
{
"epoch": 0.25,
"eval_accuracy": 0.5714285714285714,
"eval_loss": 1.1783339977264404,
"eval_runtime": 4.1774,
"eval_samples_per_second": 16.757,
"eval_steps_per_second": 16.757,
"step": 300
},
{
"epoch": 1.01,
"learning_rate": 4.1203703703703705e-05,
"loss": 0.7829,
"step": 310
},
{
"epoch": 1.02,
"learning_rate": 4.074074074074074e-05,
"loss": 0.744,
"step": 320
},
{
"epoch": 1.02,
"learning_rate": 4.027777777777778e-05,
"loss": 1.6876,
"step": 330
},
{
"epoch": 1.03,
"learning_rate": 3.981481481481482e-05,
"loss": 1.6604,
"step": 340
},
{
"epoch": 1.04,
"learning_rate": 3.935185185185186e-05,
"loss": 0.9806,
"step": 350
},
{
"epoch": 1.05,
"learning_rate": 3.888888888888889e-05,
"loss": 1.3507,
"step": 360
},
{
"epoch": 1.06,
"learning_rate": 3.8425925925925924e-05,
"loss": 1.6296,
"step": 370
},
{
"epoch": 1.07,
"learning_rate": 3.7962962962962964e-05,
"loss": 1.7596,
"step": 380
},
{
"epoch": 1.07,
"learning_rate": 3.7500000000000003e-05,
"loss": 1.3355,
"step": 390
},
{
"epoch": 1.08,
"learning_rate": 3.7037037037037037e-05,
"loss": 1.1688,
"step": 400
},
{
"epoch": 1.09,
"learning_rate": 3.6574074074074076e-05,
"loss": 0.3155,
"step": 410
},
{
"epoch": 1.1,
"learning_rate": 3.611111111111111e-05,
"loss": 1.5077,
"step": 420
},
{
"epoch": 1.11,
"learning_rate": 3.564814814814815e-05,
"loss": 1.4739,
"step": 430
},
{
"epoch": 1.12,
"learning_rate": 3.518518518518519e-05,
"loss": 1.0001,
"step": 440
},
{
"epoch": 1.12,
"learning_rate": 3.472222222222222e-05,
"loss": 0.7747,
"step": 450
},
{
"epoch": 1.13,
"learning_rate": 3.425925925925926e-05,
"loss": 0.7623,
"step": 460
},
{
"epoch": 1.14,
"learning_rate": 3.3796296296296295e-05,
"loss": 0.5152,
"step": 470
},
{
"epoch": 1.15,
"learning_rate": 3.3333333333333335e-05,
"loss": 0.8586,
"step": 480
},
{
"epoch": 1.16,
"learning_rate": 3.2870370370370375e-05,
"loss": 0.829,
"step": 490
},
{
"epoch": 1.17,
"learning_rate": 3.240740740740741e-05,
"loss": 0.925,
"step": 500
},
{
"epoch": 1.18,
"learning_rate": 3.194444444444444e-05,
"loss": 0.8935,
"step": 510
},
{
"epoch": 1.18,
"learning_rate": 3.148148148148148e-05,
"loss": 0.043,
"step": 520
},
{
"epoch": 1.19,
"learning_rate": 3.101851851851852e-05,
"loss": 0.8063,
"step": 530
},
{
"epoch": 1.2,
"learning_rate": 3.055555555555556e-05,
"loss": 0.6891,
"step": 540
},
{
"epoch": 1.21,
"learning_rate": 3.0092592592592593e-05,
"loss": 0.1838,
"step": 550
},
{
"epoch": 1.22,
"learning_rate": 2.962962962962963e-05,
"loss": 0.074,
"step": 560
},
{
"epoch": 1.23,
"learning_rate": 2.916666666666667e-05,
"loss": 1.198,
"step": 570
},
{
"epoch": 1.23,
"learning_rate": 2.8703703703703706e-05,
"loss": 1.4145,
"step": 580
},
{
"epoch": 1.24,
"learning_rate": 2.824074074074074e-05,
"loss": 0.8446,
"step": 590
},
{
"epoch": 1.25,
"learning_rate": 2.777777777777778e-05,
"loss": 2.5838,
"step": 600
},
{
"epoch": 1.25,
"eval_accuracy": 0.6,
"eval_loss": 1.3998225927352905,
"eval_runtime": 4.229,
"eval_samples_per_second": 16.552,
"eval_steps_per_second": 16.552,
"step": 600
},
{
"epoch": 2.01,
"learning_rate": 2.7314814814814816e-05,
"loss": 0.5016,
"step": 610
},
{
"epoch": 2.02,
"learning_rate": 2.6851851851851855e-05,
"loss": 0.1834,
"step": 620
},
{
"epoch": 2.02,
"learning_rate": 2.6388888888888892e-05,
"loss": 0.0938,
"step": 630
},
{
"epoch": 2.03,
"learning_rate": 2.5925925925925925e-05,
"loss": 0.8475,
"step": 640
},
{
"epoch": 2.04,
"learning_rate": 2.5462962962962965e-05,
"loss": 0.2572,
"step": 650
},
{
"epoch": 2.05,
"learning_rate": 2.5e-05,
"loss": 0.3269,
"step": 660
},
{
"epoch": 2.06,
"learning_rate": 2.4537037037037038e-05,
"loss": 0.5216,
"step": 670
},
{
"epoch": 2.07,
"learning_rate": 2.4074074074074074e-05,
"loss": 0.3262,
"step": 680
},
{
"epoch": 2.08,
"learning_rate": 2.361111111111111e-05,
"loss": 0.3112,
"step": 690
},
{
"epoch": 2.08,
"learning_rate": 2.314814814814815e-05,
"loss": 0.3168,
"step": 700
},
{
"epoch": 2.09,
"learning_rate": 2.2685185185185187e-05,
"loss": 0.1338,
"step": 710
},
{
"epoch": 2.1,
"learning_rate": 2.2222222222222223e-05,
"loss": 0.5042,
"step": 720
},
{
"epoch": 2.11,
"learning_rate": 2.175925925925926e-05,
"loss": 0.4495,
"step": 730
},
{
"epoch": 2.12,
"learning_rate": 2.1296296296296296e-05,
"loss": 0.6722,
"step": 740
},
{
"epoch": 2.12,
"learning_rate": 2.0833333333333336e-05,
"loss": 0.1336,
"step": 750
},
{
"epoch": 2.13,
"learning_rate": 2.037037037037037e-05,
"loss": 0.0143,
"step": 760
},
{
"epoch": 2.14,
"learning_rate": 1.990740740740741e-05,
"loss": 0.2984,
"step": 770
},
{
"epoch": 2.15,
"learning_rate": 1.9444444444444445e-05,
"loss": 0.3383,
"step": 780
},
{
"epoch": 2.16,
"learning_rate": 1.8981481481481482e-05,
"loss": 0.4476,
"step": 790
},
{
"epoch": 2.17,
"learning_rate": 1.8518518518518518e-05,
"loss": 0.3041,
"step": 800
},
{
"epoch": 2.17,
"learning_rate": 1.8055555555555555e-05,
"loss": 0.0096,
"step": 810
},
{
"epoch": 2.18,
"learning_rate": 1.7592592592592595e-05,
"loss": 0.0711,
"step": 820
},
{
"epoch": 2.19,
"learning_rate": 1.712962962962963e-05,
"loss": 0.0138,
"step": 830
},
{
"epoch": 2.2,
"learning_rate": 1.6666666666666667e-05,
"loss": 0.7224,
"step": 840
},
{
"epoch": 2.21,
"learning_rate": 1.6203703703703704e-05,
"loss": 0.3122,
"step": 850
},
{
"epoch": 2.22,
"learning_rate": 1.574074074074074e-05,
"loss": 0.0469,
"step": 860
},
{
"epoch": 2.23,
"learning_rate": 1.527777777777778e-05,
"loss": 0.5039,
"step": 870
},
{
"epoch": 2.23,
"learning_rate": 1.4814814814814815e-05,
"loss": 0.0058,
"step": 880
},
{
"epoch": 2.24,
"learning_rate": 1.4351851851851853e-05,
"loss": 0.0123,
"step": 890
},
{
"epoch": 2.25,
"learning_rate": 1.388888888888889e-05,
"loss": 0.0371,
"step": 900
},
{
"epoch": 2.25,
"eval_accuracy": 0.8,
"eval_loss": 0.9554027915000916,
"eval_runtime": 4.3481,
"eval_samples_per_second": 16.099,
"eval_steps_per_second": 16.099,
"step": 900
},
{
"epoch": 3.01,
"learning_rate": 1.3425925925925928e-05,
"loss": 0.0159,
"step": 910
},
{
"epoch": 3.02,
"learning_rate": 1.2962962962962962e-05,
"loss": 0.0044,
"step": 920
},
{
"epoch": 3.02,
"learning_rate": 1.25e-05,
"loss": 0.0141,
"step": 930
},
{
"epoch": 3.03,
"learning_rate": 1.2037037037037037e-05,
"loss": 0.0089,
"step": 940
},
{
"epoch": 3.04,
"learning_rate": 1.1574074074074075e-05,
"loss": 0.0043,
"step": 950
},
{
"epoch": 3.05,
"learning_rate": 1.1111111111111112e-05,
"loss": 0.0495,
"step": 960
},
{
"epoch": 3.06,
"learning_rate": 1.0648148148148148e-05,
"loss": 0.0486,
"step": 970
},
{
"epoch": 3.07,
"learning_rate": 1.0185185185185185e-05,
"loss": 0.008,
"step": 980
},
{
"epoch": 3.08,
"learning_rate": 9.722222222222223e-06,
"loss": 0.0102,
"step": 990
},
{
"epoch": 3.08,
"learning_rate": 9.259259259259259e-06,
"loss": 0.4365,
"step": 1000
},
{
"epoch": 3.09,
"learning_rate": 8.796296296296297e-06,
"loss": 0.0097,
"step": 1010
},
{
"epoch": 3.1,
"learning_rate": 8.333333333333334e-06,
"loss": 0.0069,
"step": 1020
},
{
"epoch": 3.11,
"learning_rate": 7.87037037037037e-06,
"loss": 0.0269,
"step": 1030
},
{
"epoch": 3.12,
"learning_rate": 7.4074074074074075e-06,
"loss": 0.018,
"step": 1040
},
{
"epoch": 3.12,
"learning_rate": 6.944444444444445e-06,
"loss": 0.2271,
"step": 1050
},
{
"epoch": 3.13,
"learning_rate": 6.481481481481481e-06,
"loss": 0.2312,
"step": 1060
},
{
"epoch": 3.14,
"learning_rate": 6.0185185185185185e-06,
"loss": 0.6102,
"step": 1070
},
{
"epoch": 3.15,
"learning_rate": 5.555555555555556e-06,
"loss": 0.0144,
"step": 1080
},
{
"epoch": 3.16,
"learning_rate": 5.092592592592592e-06,
"loss": 0.597,
"step": 1090
},
{
"epoch": 3.17,
"learning_rate": 4.6296296296296296e-06,
"loss": 0.0035,
"step": 1100
},
{
"epoch": 3.17,
"learning_rate": 4.166666666666667e-06,
"loss": 0.006,
"step": 1110
},
{
"epoch": 3.18,
"learning_rate": 3.7037037037037037e-06,
"loss": 0.0168,
"step": 1120
},
{
"epoch": 3.19,
"learning_rate": 3.2407407407407406e-06,
"loss": 0.0025,
"step": 1130
},
{
"epoch": 3.2,
"learning_rate": 2.777777777777778e-06,
"loss": 0.5126,
"step": 1140
},
{
"epoch": 3.21,
"learning_rate": 2.3148148148148148e-06,
"loss": 0.0078,
"step": 1150
},
{
"epoch": 3.22,
"learning_rate": 1.8518518518518519e-06,
"loss": 0.0037,
"step": 1160
},
{
"epoch": 3.23,
"learning_rate": 1.388888888888889e-06,
"loss": 0.0588,
"step": 1170
},
{
"epoch": 3.23,
"learning_rate": 9.259259259259259e-07,
"loss": 0.0044,
"step": 1180
},
{
"epoch": 3.24,
"learning_rate": 4.6296296296296297e-07,
"loss": 0.0038,
"step": 1190
},
{
"epoch": 3.25,
"learning_rate": 0.0,
"loss": 0.9005,
"step": 1200
},
{
"epoch": 3.25,
"eval_accuracy": 0.9428571428571428,
"eval_loss": 0.20906202495098114,
"eval_runtime": 4.249,
"eval_samples_per_second": 16.474,
"eval_steps_per_second": 16.474,
"step": 1200
},
{
"epoch": 3.25,
"step": 1200,
"total_flos": 1.495384188125184e+18,
"train_loss": 0.8461190654104576,
"train_runtime": 199.2573,
"train_samples_per_second": 6.022,
"train_steps_per_second": 6.022
},
{
"epoch": 3.25,
"eval_accuracy": 0.9741935483870968,
"eval_loss": 0.12710630893707275,
"eval_runtime": 9.8293,
"eval_samples_per_second": 15.769,
"eval_steps_per_second": 15.769,
"step": 1200
}
],
"max_steps": 1200,
"num_train_epochs": 9223372036854775807,
"total_flos": 1.495384188125184e+18,
"trial_name": null,
"trial_params": null
}