MrFazer's picture
Model save
bba506e verified
{
"best_metric": 0.8714285714285714,
"best_model_checkpoint": "videomae-base-finetuned-ucf101-subset\\checkpoint-148",
"epoch": 3.22972972972973,
"eval_steps": 500,
"global_step": 148,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.07,
"learning_rate": 3.3333333333333335e-05,
"loss": 6.1112,
"step": 10
},
{
"epoch": 0.14,
"learning_rate": 4.81203007518797e-05,
"loss": 5.4982,
"step": 20
},
{
"epoch": 0.2,
"learning_rate": 4.43609022556391e-05,
"loss": 4.5418,
"step": 30
},
{
"epoch": 0.26,
"eval_accuracy": 0.44285714285714284,
"eval_loss": 3.4915428161621094,
"eval_runtime": 114.5117,
"eval_samples_per_second": 0.611,
"eval_steps_per_second": 0.079,
"step": 38
},
{
"epoch": 1.01,
"learning_rate": 4.0601503759398494e-05,
"loss": 3.8179,
"step": 40
},
{
"epoch": 1.08,
"learning_rate": 3.6842105263157895e-05,
"loss": 3.2454,
"step": 50
},
{
"epoch": 1.15,
"learning_rate": 3.3082706766917295e-05,
"loss": 2.834,
"step": 60
},
{
"epoch": 1.22,
"learning_rate": 2.9323308270676693e-05,
"loss": 2.3042,
"step": 70
},
{
"epoch": 1.26,
"eval_accuracy": 0.6428571428571429,
"eval_loss": 2.083695888519287,
"eval_runtime": 114.5262,
"eval_samples_per_second": 0.611,
"eval_steps_per_second": 0.079,
"step": 76
},
{
"epoch": 2.03,
"learning_rate": 2.556390977443609e-05,
"loss": 2.0549,
"step": 80
},
{
"epoch": 2.09,
"learning_rate": 2.1804511278195487e-05,
"loss": 1.6432,
"step": 90
},
{
"epoch": 2.16,
"learning_rate": 1.8045112781954888e-05,
"loss": 1.4534,
"step": 100
},
{
"epoch": 2.23,
"learning_rate": 1.4285714285714285e-05,
"loss": 1.2814,
"step": 110
},
{
"epoch": 2.26,
"eval_accuracy": 0.7714285714285715,
"eval_loss": 1.2625136375427246,
"eval_runtime": 116.0788,
"eval_samples_per_second": 0.603,
"eval_steps_per_second": 0.078,
"step": 114
},
{
"epoch": 3.04,
"learning_rate": 1.0526315789473684e-05,
"loss": 1.0406,
"step": 120
},
{
"epoch": 3.11,
"learning_rate": 6.766917293233083e-06,
"loss": 0.9346,
"step": 130
},
{
"epoch": 3.18,
"learning_rate": 3.007518796992481e-06,
"loss": 0.8829,
"step": 140
},
{
"epoch": 3.23,
"eval_accuracy": 0.8714285714285714,
"eval_loss": 1.0487520694732666,
"eval_runtime": 113.3058,
"eval_samples_per_second": 0.618,
"eval_steps_per_second": 0.079,
"step": 148
},
{
"epoch": 3.23,
"step": 148,
"total_flos": 1.4656363477578547e+18,
"train_loss": 2.589907117792078,
"train_runtime": 5052.7726,
"train_samples_per_second": 0.234,
"train_steps_per_second": 0.029
},
{
"epoch": 3.23,
"eval_accuracy": 0.7741935483870968,
"eval_loss": 1.0857348442077637,
"eval_runtime": 256.7253,
"eval_samples_per_second": 0.604,
"eval_steps_per_second": 0.078,
"step": 148
},
{
"epoch": 3.23,
"eval_accuracy": 0.7741935483870968,
"eval_loss": 1.0857349634170532,
"eval_runtime": 253.9355,
"eval_samples_per_second": 0.61,
"eval_steps_per_second": 0.079,
"step": 148
},
{
"epoch": 3.23,
"eval_accuracy": 0.7741935483870968,
"eval_loss": 1.0857348442077637,
"eval_runtime": 248.5906,
"eval_samples_per_second": 0.624,
"eval_steps_per_second": 0.08,
"step": 148
},
{
"epoch": 3.23,
"eval_accuracy": 0.7741935483870968,
"eval_loss": 1.0857348442077637,
"eval_runtime": 250.3156,
"eval_samples_per_second": 0.619,
"eval_steps_per_second": 0.08,
"step": 148
},
{
"epoch": 3.23,
"eval_accuracy": 0.7741935483870968,
"eval_loss": 1.0857348442077637,
"eval_runtime": 253.7105,
"eval_samples_per_second": 0.611,
"eval_steps_per_second": 0.079,
"step": 148
}
],
"logging_steps": 10,
"max_steps": 148,
"num_train_epochs": 9223372036854775807,
"save_steps": 500,
"total_flos": 1.4656363477578547e+18,
"trial_name": null,
"trial_params": null
}