trainer_state.json

{
  "best_metric": 0.7333333333333333,
  "best_model_checkpoint": "beit-base-patch16-224-pt22k-ft22k/checkpoint-8",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 24,
  "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_loss": 3.1590805053710938,
      "eval_runtime": 3.4474,
      "eval_samples_per_second": 69.617,
      "eval_steps_per_second": 2.321,
      "step": 8
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.03333333333333333,
      "loss": 5.12,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.26666666666666666,
      "eval_loss": 2.633744716644287,
      "eval_runtime": 3.2534,
      "eval_samples_per_second": 73.768,
      "eval_steps_per_second": 2.459,
      "step": 16
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.009523809523809525,
      "loss": 1.8305,
      "step": 20
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7333333333333333,
      "eval_loss": 0.6371921300888062,
      "eval_runtime": 3.6339,
      "eval_samples_per_second": 66.045,
      "eval_steps_per_second": 2.202,
      "step": 24
    },
    {
      "epoch": 3.0,
      "step": 24,
      "total_flos": 2.32376605913088e+17,
      "train_loss": 3.054683963457743,
      "train_runtime": 175.6001,
      "train_samples_per_second": 17.084,
      "train_steps_per_second": 0.137
    }
  ],
  "logging_steps": 10,
  "max_steps": 24,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 2.32376605913088e+17,
  "trial_name": null,
  "trial_params": null
}