vit-base-patch16-224 / trainer_state.json
HorcruxNo13's picture
Training in progress, epoch 1
2ef3cb0
raw
history blame
6.96 kB
{
"best_metric": 0.9083333333333333,
"best_model_checkpoint": "vit-base-patch16-224/checkpoint-40",
"epoch": 15.0,
"eval_steps": 500,
"global_step": 60,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_accuracy": 0.8666666666666667,
"eval_f1_score": 0.8224224224224224,
"eval_loss": 0.5036832094192505,
"eval_precision": 0.8150354609929079,
"eval_recall": 0.8666666666666667,
"eval_runtime": 4.0852,
"eval_samples_per_second": 58.748,
"eval_steps_per_second": 0.979,
"step": 4
},
{
"epoch": 2.0,
"eval_accuracy": 0.8708333333333333,
"eval_f1_score": 0.8107089829250185,
"eval_loss": 0.35003846883773804,
"eval_precision": 0.7583506944444445,
"eval_recall": 0.8708333333333333,
"eval_runtime": 4.1025,
"eval_samples_per_second": 58.501,
"eval_steps_per_second": 0.975,
"step": 8
},
{
"epoch": 3.0,
"eval_accuracy": 0.8708333333333333,
"eval_f1_score": 0.8107089829250185,
"eval_loss": 0.3154493272304535,
"eval_precision": 0.7583506944444445,
"eval_recall": 0.8708333333333333,
"eval_runtime": 3.9805,
"eval_samples_per_second": 60.295,
"eval_steps_per_second": 1.005,
"step": 12
},
{
"epoch": 3.75,
"learning_rate": 4.166666666666667e-05,
"loss": 0.5284,
"step": 15
},
{
"epoch": 4.0,
"eval_accuracy": 0.8833333333333333,
"eval_f1_score": 0.849658648884655,
"eval_loss": 0.2973836362361908,
"eval_precision": 0.8659207030451664,
"eval_recall": 0.8833333333333333,
"eval_runtime": 3.9445,
"eval_samples_per_second": 60.844,
"eval_steps_per_second": 1.014,
"step": 16
},
{
"epoch": 5.0,
"eval_accuracy": 0.8875,
"eval_f1_score": 0.8768099547511312,
"eval_loss": 0.2953941226005554,
"eval_precision": 0.873125,
"eval_recall": 0.8875,
"eval_runtime": 4.0614,
"eval_samples_per_second": 59.092,
"eval_steps_per_second": 0.985,
"step": 20
},
{
"epoch": 6.0,
"eval_accuracy": 0.8958333333333334,
"eval_f1_score": 0.8716479619238106,
"eval_loss": 0.2720634639263153,
"eval_precision": 0.8870833333333333,
"eval_recall": 0.8958333333333334,
"eval_runtime": 4.0196,
"eval_samples_per_second": 59.708,
"eval_steps_per_second": 0.995,
"step": 24
},
{
"epoch": 7.0,
"eval_accuracy": 0.8875,
"eval_f1_score": 0.852667622475749,
"eval_loss": 0.26786690950393677,
"eval_precision": 0.881712962962963,
"eval_recall": 0.8875,
"eval_runtime": 4.2691,
"eval_samples_per_second": 56.218,
"eval_steps_per_second": 0.937,
"step": 28
},
{
"epoch": 7.5,
"learning_rate": 2.777777777777778e-05,
"loss": 0.3362,
"step": 30
},
{
"epoch": 8.0,
"eval_accuracy": 0.8875,
"eval_f1_score": 0.852667622475749,
"eval_loss": 0.26340213418006897,
"eval_precision": 0.881712962962963,
"eval_recall": 0.8875,
"eval_runtime": 4.2103,
"eval_samples_per_second": 57.002,
"eval_steps_per_second": 0.95,
"step": 32
},
{
"epoch": 9.0,
"eval_accuracy": 0.9041666666666667,
"eval_f1_score": 0.8879374201787995,
"eval_loss": 0.25068628787994385,
"eval_precision": 0.8952565318162663,
"eval_recall": 0.9041666666666667,
"eval_runtime": 3.9517,
"eval_samples_per_second": 60.733,
"eval_steps_per_second": 1.012,
"step": 36
},
{
"epoch": 10.0,
"eval_accuracy": 0.9083333333333333,
"eval_f1_score": 0.8940810124891471,
"eval_loss": 0.24386708438396454,
"eval_precision": 0.9006296296296298,
"eval_recall": 0.9083333333333333,
"eval_runtime": 3.9449,
"eval_samples_per_second": 60.839,
"eval_steps_per_second": 1.014,
"step": 40
},
{
"epoch": 11.0,
"eval_accuracy": 0.8916666666666667,
"eval_f1_score": 0.8884444626028217,
"eval_loss": 0.25889885425567627,
"eval_precision": 0.8860502521300643,
"eval_recall": 0.8916666666666667,
"eval_runtime": 4.1005,
"eval_samples_per_second": 58.529,
"eval_steps_per_second": 0.975,
"step": 44
},
{
"epoch": 11.25,
"learning_rate": 1.388888888888889e-05,
"loss": 0.3017,
"step": 45
},
{
"epoch": 12.0,
"eval_accuracy": 0.9083333333333333,
"eval_f1_score": 0.9024039297513475,
"eval_loss": 0.24283821880817413,
"eval_precision": 0.9004925532625392,
"eval_recall": 0.9083333333333333,
"eval_runtime": 4.2192,
"eval_samples_per_second": 56.883,
"eval_steps_per_second": 0.948,
"step": 48
},
{
"epoch": 13.0,
"eval_accuracy": 0.9,
"eval_f1_score": 0.8970256577872201,
"eval_loss": 0.2543129324913025,
"eval_precision": 0.894922622152669,
"eval_recall": 0.9,
"eval_runtime": 4.181,
"eval_samples_per_second": 57.402,
"eval_steps_per_second": 0.957,
"step": 52
},
{
"epoch": 14.0,
"eval_accuracy": 0.8958333333333334,
"eval_f1_score": 0.8951038120948916,
"eval_loss": 0.2651337683200836,
"eval_precision": 0.8944246031746032,
"eval_recall": 0.8958333333333334,
"eval_runtime": 3.9408,
"eval_samples_per_second": 60.902,
"eval_steps_per_second": 1.015,
"step": 56
},
{
"epoch": 15.0,
"learning_rate": 0.0,
"loss": 0.278,
"step": 60
},
{
"epoch": 15.0,
"eval_accuracy": 0.8958333333333334,
"eval_f1_score": 0.8951038120948916,
"eval_loss": 0.2636790871620178,
"eval_precision": 0.8944246031746032,
"eval_recall": 0.8958333333333334,
"eval_runtime": 4.1679,
"eval_samples_per_second": 57.584,
"eval_steps_per_second": 0.96,
"step": 60
},
{
"epoch": 15.0,
"step": 60,
"total_flos": 1.16237984421888e+18,
"train_loss": 0.3610760450363159,
"train_runtime": 974.8105,
"train_samples_per_second": 15.388,
"train_steps_per_second": 0.062
},
{
"epoch": 15.0,
"eval_accuracy": 0.9033333333333333,
"eval_f1_score": 0.888913443830571,
"eval_loss": 0.24460919201374054,
"eval_precision": 0.892075919335706,
"eval_recall": 0.9033333333333333,
"eval_runtime": 4.8461,
"eval_samples_per_second": 61.906,
"eval_steps_per_second": 1.032,
"step": 60
}
],
"logging_steps": 15,
"max_steps": 60,
"num_train_epochs": 15,
"save_steps": 500,
"total_flos": 1.16237984421888e+18,
"trial_name": null,
"trial_params": null
}