trainer_state.json

{
  "best_metric": 0.8248,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0360/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": 0.0005,
      "loss": 0.8075,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8026666666666666,
      "eval_loss": 0.6638714075088501,
      "eval_runtime": 12.4334,
      "eval_samples_per_second": 301.608,
      "eval_steps_per_second": 4.745,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 8.426472663879395,
      "learning_rate": 0.0005,
      "loss": 0.4568,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8064,
      "eval_loss": 0.6830059885978699,
      "eval_runtime": 11.3059,
      "eval_samples_per_second": 331.686,
      "eval_steps_per_second": 5.219,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 15.772550582885742,
      "learning_rate": 0.0005,
      "loss": 0.3412,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8176,
      "eval_loss": 0.6515712141990662,
      "eval_runtime": 13.6277,
      "eval_samples_per_second": 275.174,
      "eval_steps_per_second": 4.329,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.0397294759750366,
      "learning_rate": 0.0005,
      "loss": 0.2792,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8050666666666667,
      "eval_loss": 0.7389880418777466,
      "eval_runtime": 12.2519,
      "eval_samples_per_second": 306.074,
      "eval_steps_per_second": 4.816,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.5394655466079712,
      "learning_rate": 0.0005,
      "loss": 0.2345,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8248,
      "eval_loss": 0.6715890765190125,
      "eval_runtime": 11.9364,
      "eval_samples_per_second": 314.164,
      "eval_steps_per_second": 4.943,
      "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
}