swin-tiny-patch4-window7-224 / trainer_state.json
HorcruxNo13's picture
Training in progress, epoch 1
31284f4
raw
history blame
4.16 kB
{
"best_metric": 0.7625,
"best_model_checkpoint": "swin-tiny-patch4-window7-224/checkpoint-48",
"epoch": 7.0,
"eval_steps": 500,
"global_step": 56,
"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": 0.6205128205128204,
"eval_loss": 0.5696213841438293,
"eval_precision": 0.5377777777777778,
"eval_recall": 0.7333333333333333,
"eval_runtime": 1.8813,
"eval_samples_per_second": 127.574,
"eval_steps_per_second": 4.252,
"step": 8
},
{
"epoch": 1.25,
"learning_rate": 4.600000000000001e-05,
"loss": 0.6683,
"step": 10
},
{
"epoch": 2.0,
"eval_accuracy": 0.7333333333333333,
"eval_f1_score": 0.6205128205128204,
"eval_loss": 0.5635157227516174,
"eval_precision": 0.5377777777777778,
"eval_recall": 0.7333333333333333,
"eval_runtime": 1.8341,
"eval_samples_per_second": 130.853,
"eval_steps_per_second": 4.362,
"step": 16
},
{
"epoch": 2.5,
"learning_rate": 3.6e-05,
"loss": 0.5797,
"step": 20
},
{
"epoch": 3.0,
"eval_accuracy": 0.7333333333333333,
"eval_f1_score": 0.6205128205128204,
"eval_loss": 0.55841064453125,
"eval_precision": 0.5377777777777778,
"eval_recall": 0.7333333333333333,
"eval_runtime": 2.0336,
"eval_samples_per_second": 118.017,
"eval_steps_per_second": 3.934,
"step": 24
},
{
"epoch": 3.75,
"learning_rate": 2.6000000000000002e-05,
"loss": 0.5547,
"step": 30
},
{
"epoch": 4.0,
"eval_accuracy": 0.7333333333333333,
"eval_f1_score": 0.6205128205128204,
"eval_loss": 0.5731536746025085,
"eval_precision": 0.5377777777777778,
"eval_recall": 0.7333333333333333,
"eval_runtime": 1.8352,
"eval_samples_per_second": 130.778,
"eval_steps_per_second": 4.359,
"step": 32
},
{
"epoch": 5.0,
"learning_rate": 1.6000000000000003e-05,
"loss": 0.5165,
"step": 40
},
{
"epoch": 5.0,
"eval_accuracy": 0.7583333333333333,
"eval_f1_score": 0.6959050899349406,
"eval_loss": 0.5416003465652466,
"eval_precision": 0.7485882848714707,
"eval_recall": 0.7583333333333333,
"eval_runtime": 2.091,
"eval_samples_per_second": 114.78,
"eval_steps_per_second": 3.826,
"step": 40
},
{
"epoch": 6.0,
"eval_accuracy": 0.7625,
"eval_f1_score": 0.7033555352126014,
"eval_loss": 0.5488410592079163,
"eval_precision": 0.7561481481481481,
"eval_recall": 0.7625,
"eval_runtime": 2.0904,
"eval_samples_per_second": 114.812,
"eval_steps_per_second": 3.827,
"step": 48
},
{
"epoch": 6.25,
"learning_rate": 6e-06,
"loss": 0.4893,
"step": 50
},
{
"epoch": 7.0,
"eval_accuracy": 0.7583333333333333,
"eval_f1_score": 0.7003333333333333,
"eval_loss": 0.5511656403541565,
"eval_precision": 0.7431547619047619,
"eval_recall": 0.7583333333333333,
"eval_runtime": 1.9003,
"eval_samples_per_second": 126.293,
"eval_steps_per_second": 4.21,
"step": 56
},
{
"epoch": 7.0,
"step": 56,
"total_flos": 1.73991922900992e+17,
"train_loss": 0.5527358267988477,
"train_runtime": 138.7489,
"train_samples_per_second": 50.451,
"train_steps_per_second": 0.404
},
{
"epoch": 7.0,
"eval_accuracy": 0.78,
"eval_f1_score": 0.7315167493796526,
"eval_loss": 0.5118212699890137,
"eval_precision": 0.7896499764558155,
"eval_recall": 0.78,
"eval_runtime": 2.3258,
"eval_samples_per_second": 128.99,
"eval_steps_per_second": 4.3,
"step": 56
}
],
"logging_steps": 10,
"max_steps": 56,
"num_train_epochs": 7,
"save_steps": 500,
"total_flos": 1.73991922900992e+17,
"trial_name": null,
"trial_params": null
}