|
{
|
|
"best_metric": 0.7333333333333333,
|
|
"best_model_checkpoint": "vit-base-patch16-224-RU5-10\\checkpoint-87",
|
|
"epoch": 9.23076923076923,
|
|
"eval_steps": 500,
|
|
"global_step": 90,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"epoch": 0.92,
|
|
"eval_accuracy": 0.4666666666666667,
|
|
"eval_loss": 1.293900728225708,
|
|
"eval_runtime": 1.019,
|
|
"eval_samples_per_second": 58.879,
|
|
"eval_steps_per_second": 1.963,
|
|
"step": 9
|
|
},
|
|
{
|
|
"epoch": 1.03,
|
|
"learning_rate": 5.1764705882352943e-05,
|
|
"loss": 1.3501,
|
|
"step": 10
|
|
},
|
|
{
|
|
"epoch": 1.95,
|
|
"eval_accuracy": 0.5833333333333334,
|
|
"eval_loss": 1.170591115951538,
|
|
"eval_runtime": 1.0727,
|
|
"eval_samples_per_second": 55.933,
|
|
"eval_steps_per_second": 1.864,
|
|
"step": 19
|
|
},
|
|
{
|
|
"epoch": 2.05,
|
|
"learning_rate": 4.5294117647058826e-05,
|
|
"loss": 1.2272,
|
|
"step": 20
|
|
},
|
|
{
|
|
"epoch": 2.97,
|
|
"eval_accuracy": 0.6333333333333333,
|
|
"eval_loss": 1.0594052076339722,
|
|
"eval_runtime": 1.0633,
|
|
"eval_samples_per_second": 56.426,
|
|
"eval_steps_per_second": 1.881,
|
|
"step": 29
|
|
},
|
|
{
|
|
"epoch": 3.08,
|
|
"learning_rate": 3.882352941176471e-05,
|
|
"loss": 1.0941,
|
|
"step": 30
|
|
},
|
|
{
|
|
"epoch": 4.0,
|
|
"eval_accuracy": 0.6,
|
|
"eval_loss": 0.9773460030555725,
|
|
"eval_runtime": 1.0306,
|
|
"eval_samples_per_second": 58.221,
|
|
"eval_steps_per_second": 1.941,
|
|
"step": 39
|
|
},
|
|
{
|
|
"epoch": 4.1,
|
|
"learning_rate": 3.235294117647059e-05,
|
|
"loss": 0.979,
|
|
"step": 40
|
|
},
|
|
{
|
|
"epoch": 4.92,
|
|
"eval_accuracy": 0.6833333333333333,
|
|
"eval_loss": 0.9141623377799988,
|
|
"eval_runtime": 1.0244,
|
|
"eval_samples_per_second": 58.572,
|
|
"eval_steps_per_second": 1.952,
|
|
"step": 48
|
|
},
|
|
{
|
|
"epoch": 5.13,
|
|
"learning_rate": 2.5882352941176472e-05,
|
|
"loss": 0.8694,
|
|
"step": 50
|
|
},
|
|
{
|
|
"epoch": 5.95,
|
|
"eval_accuracy": 0.7,
|
|
"eval_loss": 0.8568627238273621,
|
|
"eval_runtime": 1.0443,
|
|
"eval_samples_per_second": 57.453,
|
|
"eval_steps_per_second": 1.915,
|
|
"step": 58
|
|
},
|
|
{
|
|
"epoch": 6.15,
|
|
"learning_rate": 1.9411764705882355e-05,
|
|
"loss": 0.7662,
|
|
"step": 60
|
|
},
|
|
{
|
|
"epoch": 6.97,
|
|
"eval_accuracy": 0.6833333333333333,
|
|
"eval_loss": 0.8363937139511108,
|
|
"eval_runtime": 1.067,
|
|
"eval_samples_per_second": 56.233,
|
|
"eval_steps_per_second": 1.874,
|
|
"step": 68
|
|
},
|
|
{
|
|
"epoch": 7.18,
|
|
"learning_rate": 1.2941176470588236e-05,
|
|
"loss": 0.7002,
|
|
"step": 70
|
|
},
|
|
{
|
|
"epoch": 8.0,
|
|
"eval_accuracy": 0.7,
|
|
"eval_loss": 0.8071037530899048,
|
|
"eval_runtime": 1.0561,
|
|
"eval_samples_per_second": 56.814,
|
|
"eval_steps_per_second": 1.894,
|
|
"step": 78
|
|
},
|
|
{
|
|
"epoch": 8.21,
|
|
"learning_rate": 6.470588235294118e-06,
|
|
"loss": 0.6443,
|
|
"step": 80
|
|
},
|
|
{
|
|
"epoch": 8.92,
|
|
"eval_accuracy": 0.7333333333333333,
|
|
"eval_loss": 0.809493362903595,
|
|
"eval_runtime": 1.0229,
|
|
"eval_samples_per_second": 58.655,
|
|
"eval_steps_per_second": 1.955,
|
|
"step": 87
|
|
},
|
|
{
|
|
"epoch": 9.23,
|
|
"learning_rate": 0.0,
|
|
"loss": 0.629,
|
|
"step": 90
|
|
},
|
|
{
|
|
"epoch": 9.23,
|
|
"eval_accuracy": 0.7166666666666667,
|
|
"eval_loss": 0.8133583664894104,
|
|
"eval_runtime": 1.0093,
|
|
"eval_samples_per_second": 59.45,
|
|
"eval_steps_per_second": 1.982,
|
|
"step": 90
|
|
},
|
|
{
|
|
"epoch": 9.23,
|
|
"step": 90,
|
|
"total_flos": 8.766825930857595e+17,
|
|
"train_loss": 0.9177195443047418,
|
|
"train_runtime": 195.1818,
|
|
"train_samples_per_second": 62.762,
|
|
"train_steps_per_second": 0.461
|
|
}
|
|
],
|
|
"logging_steps": 10,
|
|
"max_steps": 90,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"total_flos": 8.766825930857595e+17,
|
|
"train_batch_size": 32,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|
|
|