trainer_state.json

{
  "best_metric": 0.9088,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0880/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": 10.05113410949707,
      "learning_rate": 0.0004522542485937369,
      "loss": 0.7978,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8141333333333334,
      "eval_loss": 0.6658827662467957,
      "eval_runtime": 12.4099,
      "eval_samples_per_second": 302.177,
      "eval_steps_per_second": 4.754,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00032770273425644284,
      "loss": 0.4065,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8568,
      "eval_loss": 0.4801413416862488,
      "eval_runtime": 13.3803,
      "eval_samples_per_second": 280.264,
      "eval_steps_per_second": 4.409,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.027277149260044098,
      "learning_rate": 0.00017319451210748358,
      "loss": 0.1745,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8781333333333333,
      "eval_loss": 0.45083847641944885,
      "eval_runtime": 12.4598,
      "eval_samples_per_second": 300.967,
      "eval_steps_per_second": 4.735,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.0996912717819214,
      "learning_rate": 4.802337581094707e-05,
      "loss": 0.0436,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9037333333333334,
      "eval_loss": 0.37227463722229004,
      "eval_runtime": 12.0482,
      "eval_samples_per_second": 311.251,
      "eval_steps_per_second": 4.897,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.0071251592598855495,
      "learning_rate": 4.4502166460236306e-10,
      "loss": 0.0061,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9088,
      "eval_loss": 0.35868656635284424,
      "eval_runtime": 12.1896,
      "eval_samples_per_second": 307.639,
      "eval_steps_per_second": 4.84,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}