trainer_state.json

{
  "best_metric": 0.3486238532110092,
  "best_model_checkpoint": "swiftformer-xs-dmae-va-U-SF/checkpoint-21",
  "epoch": 2.709677419354839,
  "eval_steps": 500,
  "global_step": 21,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9,
      "eval_accuracy": 0.24770642201834864,
      "eval_loss": 1.4376270771026611,
      "eval_runtime": 0.5029,
      "eval_samples_per_second": 216.756,
      "eval_steps_per_second": 7.954,
      "step": 7
    },
    {
      "epoch": 1.29,
      "learning_rate": 3.055555555555556e-05,
      "loss": 1.4126,
      "step": 10
    },
    {
      "epoch": 1.94,
      "eval_accuracy": 0.3211009174311927,
      "eval_loss": 1.3581504821777344,
      "eval_runtime": 0.4287,
      "eval_samples_per_second": 254.235,
      "eval_steps_per_second": 9.33,
      "step": 15
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.777777777777778e-06,
      "loss": 1.359,
      "step": 20
    },
    {
      "epoch": 2.71,
      "eval_accuracy": 0.3486238532110092,
      "eval_loss": 1.3463064432144165,
      "eval_runtime": 0.4247,
      "eval_samples_per_second": 256.633,
      "eval_steps_per_second": 9.418,
      "step": 21
    },
    {
      "epoch": 2.71,
      "step": 21,
      "total_flos": 7291673091440640.0,
      "train_loss": 1.383760514713469,
      "train_runtime": 25.3309,
      "train_samples_per_second": 115.827,
      "train_steps_per_second": 0.829
    }
  ],
  "logging_steps": 10,
  "max_steps": 21,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 7291673091440640.0,
  "trial_name": null,
  "trial_params": null
}