Tommidi's picture
End of training
e7d5dfd verified
raw
history blame
3.87 kB
{
"best_metric": 0.7333333333333333,
"best_model_checkpoint": "spatial_vit_temporal_vit-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": 2.3018,
"step": 10
},
{
"epoch": 0.14,
"learning_rate": 4.81203007518797e-05,
"loss": 2.2636,
"step": 20
},
{
"epoch": 0.2,
"learning_rate": 4.43609022556391e-05,
"loss": 2.2165,
"step": 30
},
{
"epoch": 0.26,
"eval_accuracy": 0.26666666666666666,
"eval_loss": 2.0699431896209717,
"eval_runtime": 107.0916,
"eval_samples_per_second": 0.28,
"eval_steps_per_second": 0.037,
"step": 38
},
{
"epoch": 1.01,
"learning_rate": 4.0601503759398494e-05,
"loss": 2.1348,
"step": 40
},
{
"epoch": 1.08,
"learning_rate": 3.6842105263157895e-05,
"loss": 2.008,
"step": 50
},
{
"epoch": 1.15,
"learning_rate": 3.3082706766917295e-05,
"loss": 1.9693,
"step": 60
},
{
"epoch": 1.22,
"learning_rate": 2.9323308270676693e-05,
"loss": 1.8229,
"step": 70
},
{
"epoch": 1.26,
"eval_accuracy": 0.5,
"eval_loss": 1.816007375717163,
"eval_runtime": 108.4965,
"eval_samples_per_second": 0.277,
"eval_steps_per_second": 0.037,
"step": 76
},
{
"epoch": 2.03,
"learning_rate": 2.556390977443609e-05,
"loss": 1.8013,
"step": 80
},
{
"epoch": 2.09,
"learning_rate": 2.1804511278195487e-05,
"loss": 1.6652,
"step": 90
},
{
"epoch": 2.16,
"learning_rate": 1.8045112781954888e-05,
"loss": 1.6004,
"step": 100
},
{
"epoch": 2.23,
"learning_rate": 1.4285714285714285e-05,
"loss": 1.4707,
"step": 110
},
{
"epoch": 2.26,
"eval_accuracy": 0.7,
"eval_loss": 1.4157180786132812,
"eval_runtime": 108.1494,
"eval_samples_per_second": 0.277,
"eval_steps_per_second": 0.037,
"step": 114
},
{
"epoch": 3.04,
"learning_rate": 1.0526315789473684e-05,
"loss": 1.4613,
"step": 120
},
{
"epoch": 3.11,
"learning_rate": 6.766917293233083e-06,
"loss": 1.3109,
"step": 130
},
{
"epoch": 3.18,
"learning_rate": 3.007518796992481e-06,
"loss": 1.3886,
"step": 140
},
{
"epoch": 3.23,
"eval_accuracy": 0.7333333333333333,
"eval_loss": 1.3519634008407593,
"eval_runtime": 108.7503,
"eval_samples_per_second": 0.276,
"eval_steps_per_second": 0.037,
"step": 148
},
{
"epoch": 3.23,
"step": 148,
"total_flos": 0.0,
"train_loss": 1.7922898885366079,
"train_runtime": 5180.8771,
"train_samples_per_second": 0.229,
"train_steps_per_second": 0.029
},
{
"epoch": 3.23,
"eval_accuracy": 0.68,
"eval_loss": 1.317222237586975,
"eval_runtime": 274.8188,
"eval_samples_per_second": 0.273,
"eval_steps_per_second": 0.036,
"step": 148
},
{
"epoch": 3.23,
"eval_accuracy": 0.68,
"eval_loss": 1.317222237586975,
"eval_runtime": 271.9717,
"eval_samples_per_second": 0.276,
"eval_steps_per_second": 0.037,
"step": 148
}
],
"logging_steps": 10,
"max_steps": 148,
"num_input_tokens_seen": 0,
"num_train_epochs": 9223372036854775807,
"save_steps": 500,
"total_flos": 0.0,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}