trainer_state.json

{
  "best_metric": 0.32634032634032634,
  "best_model_checkpoint": "convnextv2-tiny-1k-224-for-pre_evaluation/checkpoint-21",
  "epoch": 2.7906976744186047,
  "eval_steps": 500,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.93,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 1.5431,
      "step": 10
    },
    {
      "epoch": 0.93,
      "eval_accuracy": 0.3146853146853147,
      "eval_loss": 1.5139713287353516,
      "eval_runtime": 5.0638,
      "eval_samples_per_second": 84.719,
      "eval_steps_per_second": 2.765,
      "step": 10
    },
    {
      "epoch": 1.86,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 1.4792,
      "step": 20
    },
    {
      "epoch": 1.95,
      "eval_accuracy": 0.32634032634032634,
      "eval_loss": 1.5076048374176025,
      "eval_runtime": 6.2058,
      "eval_samples_per_second": 69.129,
      "eval_steps_per_second": 2.256,
      "step": 21
    },
    {
      "epoch": 2.79,
      "learning_rate": 0.0,
      "loss": 1.4714,
      "step": 30
    },
    {
      "epoch": 2.79,
      "eval_accuracy": 0.32167832167832167,
      "eval_loss": 1.5027328729629517,
      "eval_runtime": 5.9368,
      "eval_samples_per_second": 72.262,
      "eval_steps_per_second": 2.358,
      "step": 30
    },
    {
      "epoch": 2.79,
      "step": 30,
      "total_flos": 9.58534533342167e+16,
      "train_loss": 1.4979065895080566,
      "train_runtime": 118.365,
      "train_samples_per_second": 34.47,
      "train_steps_per_second": 0.253
    }
  ],
  "logging_steps": 10,
  "max_steps": 30,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 9.58534533342167e+16,
  "trial_name": null,
  "trial_params": null
}