trainer_state.json

{
  "best_metric": 0.18897637795275588,
  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-mgasior/checkpoint-9",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 27,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_f1": 0.18897637795275588,
      "eval_loss": 46788898816.0,
      "eval_runtime": 2.512,
      "eval_samples_per_second": 50.558,
      "eval_steps_per_second": 1.592,
      "step": 9
    },
    {
      "epoch": 1.11,
      "learning_rate": 3.541666666666667e-05,
      "loss": 40845207142.4,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_f1": 0.1732283464566929,
      "eval_loss": 46788898816.0,
      "eval_runtime": 2.1609,
      "eval_samples_per_second": 58.772,
      "eval_steps_per_second": 1.851,
      "step": 18
    },
    {
      "epoch": 2.22,
      "learning_rate": 1.4583333333333335e-05,
      "loss": 41037873152.0,
      "step": 20
    },
    {
      "epoch": 3.0,
      "eval_f1": 0.1732283464566929,
      "eval_loss": 46788898816.0,
      "eval_runtime": 2.2498,
      "eval_samples_per_second": 56.448,
      "eval_steps_per_second": 1.778,
      "step": 27
    },
    {
      "epoch": 3.0,
      "step": 27,
      "total_flos": 8.516613618951782e+16,
      "train_loss": 41129279488.0,
      "train_runtime": 103.012,
      "train_samples_per_second": 33.258,
      "train_steps_per_second": 0.262
    }
  ],
  "logging_steps": 10,
  "max_steps": 27,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 8.516613618951782e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}