trainer_state.json

{
  "best_metric": 0.9328,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0468/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": Infinity,
      "learning_rate": 5e-05,
      "loss": 0.8928,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9189333333333334,
      "eval_loss": 0.3136594295501709,
      "eval_runtime": 12.5366,
      "eval_samples_per_second": 299.124,
      "eval_steps_per_second": 4.706,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.47008639574050903,
      "learning_rate": 5e-05,
      "loss": 0.1698,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9232,
      "eval_loss": 0.2588919699192047,
      "eval_runtime": 12.6807,
      "eval_samples_per_second": 295.726,
      "eval_steps_per_second": 4.653,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.4129537045955658,
      "learning_rate": 5e-05,
      "loss": 0.0924,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9285333333333333,
      "eval_loss": 0.25332462787628174,
      "eval_runtime": 12.499,
      "eval_samples_per_second": 300.024,
      "eval_steps_per_second": 4.72,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.049137771129608154,
      "learning_rate": 5e-05,
      "loss": 0.0597,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9290666666666667,
      "eval_loss": 0.27535176277160645,
      "eval_runtime": 12.4718,
      "eval_samples_per_second": 300.677,
      "eval_steps_per_second": 4.731,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.038216594606637955,
      "learning_rate": 5e-05,
      "loss": 0.0412,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9328,
      "eval_loss": 0.25949805974960327,
      "eval_runtime": 12.3795,
      "eval_samples_per_second": 302.92,
      "eval_steps_per_second": 4.766,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}