trainer_state.json

{
  "best_metric": 0.5833333333333334,
  "best_model_checkpoint": "vit-base-patch16-224-RU5-10\\checkpoint-16",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.48333333333333334,
      "eval_loss": 1.371477723121643,
      "eval_runtime": 1.144,
      "eval_samples_per_second": 52.447,
      "eval_steps_per_second": 1.748,
      "step": 2
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.4666666666666667,
      "eval_loss": 1.3415111303329468,
      "eval_runtime": 1.0274,
      "eval_samples_per_second": 58.399,
      "eval_steps_per_second": 1.947,
      "step": 4
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.4666666666666667,
      "eval_loss": 1.31484854221344,
      "eval_runtime": 1.0106,
      "eval_samples_per_second": 59.368,
      "eval_steps_per_second": 1.979,
      "step": 6
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.48333333333333334,
      "eval_loss": 1.2919448614120483,
      "eval_runtime": 1.0367,
      "eval_samples_per_second": 57.876,
      "eval_steps_per_second": 1.929,
      "step": 8
    },
    {
      "epoch": 5.0,
      "learning_rate": 2.894736842105263e-05,
      "loss": 1.3369,
      "step": 10
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.48333333333333334,
      "eval_loss": 1.272611141204834,
      "eval_runtime": 0.996,
      "eval_samples_per_second": 60.243,
      "eval_steps_per_second": 2.008,
      "step": 10
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5,
      "eval_loss": 1.2568950653076172,
      "eval_runtime": 1.0111,
      "eval_samples_per_second": 59.34,
      "eval_steps_per_second": 1.978,
      "step": 12
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.55,
      "eval_loss": 1.24420964717865,
      "eval_runtime": 1.0116,
      "eval_samples_per_second": 59.311,
      "eval_steps_per_second": 1.977,
      "step": 14
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5833333333333334,
      "eval_loss": 1.2348047494888306,
      "eval_runtime": 1.0128,
      "eval_samples_per_second": 59.24,
      "eval_steps_per_second": 1.975,
      "step": 16
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.5833333333333334,
      "eval_loss": 1.2286779880523682,
      "eval_runtime": 1.0097,
      "eval_samples_per_second": 59.423,
      "eval_steps_per_second": 1.981,
      "step": 18
    },
    {
      "epoch": 10.0,
      "learning_rate": 0.0,
      "loss": 1.2441,
      "step": 20
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.5666666666666667,
      "eval_loss": 1.2260597944259644,
      "eval_runtime": 1.032,
      "eval_samples_per_second": 58.141,
      "eval_steps_per_second": 1.938,
      "step": 20
    },
    {
      "epoch": 10.0,
      "step": 20,
      "total_flos": 1.859841088487424e+17,
      "train_loss": 1.2905001163482666,
      "train_runtime": 87.3836,
      "train_samples_per_second": 27.465,
      "train_steps_per_second": 0.229
    }
  ],
  "logging_steps": 10,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 1.859841088487424e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}