vit-base-patch16-224 / trainer_state.json
HorcruxNo13's picture
Training in progress, epoch 1
4db333f
raw
history blame
6.93 kB
{
"best_metric": 0.7916666666666666,
"best_model_checkpoint": "vit-base-patch16-224/checkpoint-52",
"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": 0.7333333333333333,
"eval_f1_score": 0.6280486019616454,
"eval_loss": 0.6009885668754578,
"eval_precision": 0.6725490196078432,
"eval_recall": 0.7333333333333333,
"eval_runtime": 73.0728,
"eval_samples_per_second": 3.284,
"eval_steps_per_second": 0.055,
"step": 4
},
{
"epoch": 2.0,
"eval_accuracy": 0.7375,
"eval_f1_score": 0.6302131603336423,
"eval_loss": 0.5551710724830627,
"eval_precision": 0.8066945606694561,
"eval_recall": 0.7375,
"eval_runtime": 4.1084,
"eval_samples_per_second": 58.416,
"eval_steps_per_second": 0.974,
"step": 8
},
{
"epoch": 3.0,
"eval_accuracy": 0.7541666666666667,
"eval_f1_score": 0.6781685795384426,
"eval_loss": 0.5449685454368591,
"eval_precision": 0.7597883597883598,
"eval_recall": 0.7541666666666667,
"eval_runtime": 4.0131,
"eval_samples_per_second": 59.804,
"eval_steps_per_second": 0.997,
"step": 12
},
{
"epoch": 3.75,
"learning_rate": 4.166666666666667e-05,
"loss": 0.576,
"step": 15
},
{
"epoch": 4.0,
"eval_accuracy": 0.75,
"eval_f1_score": 0.6641114982578398,
"eval_loss": 0.5325278043746948,
"eval_precision": 0.7706552706552707,
"eval_recall": 0.75,
"eval_runtime": 4.224,
"eval_samples_per_second": 56.818,
"eval_steps_per_second": 0.947,
"step": 16
},
{
"epoch": 5.0,
"eval_accuracy": 0.75,
"eval_f1_score": 0.6899999999999998,
"eval_loss": 0.5234411954879761,
"eval_precision": 0.7232142857142857,
"eval_recall": 0.75,
"eval_runtime": 3.9713,
"eval_samples_per_second": 60.434,
"eval_steps_per_second": 1.007,
"step": 20
},
{
"epoch": 6.0,
"eval_accuracy": 0.7625,
"eval_f1_score": 0.7075837742504409,
"eval_loss": 0.511196494102478,
"eval_precision": 0.7505671326826696,
"eval_recall": 0.7625,
"eval_runtime": 3.9183,
"eval_samples_per_second": 61.251,
"eval_steps_per_second": 1.021,
"step": 24
},
{
"epoch": 7.0,
"eval_accuracy": 0.7666666666666667,
"eval_f1_score": 0.7221264707039705,
"eval_loss": 0.5081896185874939,
"eval_precision": 0.7503197108701696,
"eval_recall": 0.7666666666666667,
"eval_runtime": 4.0209,
"eval_samples_per_second": 59.688,
"eval_steps_per_second": 0.995,
"step": 28
},
{
"epoch": 7.5,
"learning_rate": 2.777777777777778e-05,
"loss": 0.4876,
"step": 30
},
{
"epoch": 8.0,
"eval_accuracy": 0.7666666666666667,
"eval_f1_score": 0.7287749287749288,
"eval_loss": 0.5066744089126587,
"eval_precision": 0.746633117661155,
"eval_recall": 0.7666666666666667,
"eval_runtime": 3.9922,
"eval_samples_per_second": 60.118,
"eval_steps_per_second": 1.002,
"step": 32
},
{
"epoch": 9.0,
"eval_accuracy": 0.7791666666666667,
"eval_f1_score": 0.7528159852134981,
"eval_loss": 0.5091281533241272,
"eval_precision": 0.76231884057971,
"eval_recall": 0.7791666666666667,
"eval_runtime": 4.1183,
"eval_samples_per_second": 58.276,
"eval_steps_per_second": 0.971,
"step": 36
},
{
"epoch": 10.0,
"eval_accuracy": 0.7583333333333333,
"eval_f1_score": 0.704514442129346,
"eval_loss": 0.5022971630096436,
"eval_precision": 0.7393393393393393,
"eval_recall": 0.7583333333333333,
"eval_runtime": 3.9137,
"eval_samples_per_second": 61.323,
"eval_steps_per_second": 1.022,
"step": 40
},
{
"epoch": 11.0,
"eval_accuracy": 0.7708333333333334,
"eval_f1_score": 0.7434882865423094,
"eval_loss": 0.4910615086555481,
"eval_precision": 0.7506953593910115,
"eval_recall": 0.7708333333333334,
"eval_runtime": 3.9528,
"eval_samples_per_second": 60.716,
"eval_steps_per_second": 1.012,
"step": 44
},
{
"epoch": 11.25,
"learning_rate": 1.388888888888889e-05,
"loss": 0.4379,
"step": 45
},
{
"epoch": 12.0,
"eval_accuracy": 0.7666666666666667,
"eval_f1_score": 0.7513341298287536,
"eval_loss": 0.492078959941864,
"eval_precision": 0.7487322201607916,
"eval_recall": 0.7666666666666667,
"eval_runtime": 4.01,
"eval_samples_per_second": 59.85,
"eval_steps_per_second": 0.998,
"step": 48
},
{
"epoch": 13.0,
"eval_accuracy": 0.7916666666666666,
"eval_f1_score": 0.7679595398368771,
"eval_loss": 0.4905916154384613,
"eval_precision": 0.7791738054445079,
"eval_recall": 0.7916666666666666,
"eval_runtime": 4.0131,
"eval_samples_per_second": 59.804,
"eval_steps_per_second": 0.997,
"step": 52
},
{
"epoch": 14.0,
"eval_accuracy": 0.7875,
"eval_f1_score": 0.76446353296747,
"eval_loss": 0.49192750453948975,
"eval_precision": 0.7731010452961672,
"eval_recall": 0.7875,
"eval_runtime": 4.0774,
"eval_samples_per_second": 58.86,
"eval_steps_per_second": 0.981,
"step": 56
},
{
"epoch": 15.0,
"learning_rate": 0.0,
"loss": 0.4003,
"step": 60
},
{
"epoch": 15.0,
"eval_accuracy": 0.7833333333333333,
"eval_f1_score": 0.7586779214303522,
"eval_loss": 0.49290910363197327,
"eval_precision": 0.7677707976394442,
"eval_recall": 0.7833333333333333,
"eval_runtime": 3.978,
"eval_samples_per_second": 60.332,
"eval_steps_per_second": 1.006,
"step": 60
},
{
"epoch": 15.0,
"step": 60,
"total_flos": 1.16237984421888e+18,
"train_loss": 0.475440772374471,
"train_runtime": 1200.7807,
"train_samples_per_second": 12.492,
"train_steps_per_second": 0.05
},
{
"epoch": 15.0,
"eval_accuracy": 0.7833333333333333,
"eval_f1_score": 0.7558981348884555,
"eval_loss": 0.4771976172924042,
"eval_precision": 0.7701923076923076,
"eval_recall": 0.7833333333333333,
"eval_runtime": 93.6348,
"eval_samples_per_second": 3.204,
"eval_steps_per_second": 0.053,
"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
}