trainer_state.json

{
  "best_metric": 0.9389333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0364/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": 0.167902871966362,
      "learning_rate": 0.0001,
      "loss": 0.5479,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9330666666666667,
      "eval_loss": 0.21452213823795319,
      "eval_runtime": 6.2464,
      "eval_samples_per_second": 600.342,
      "eval_steps_per_second": 9.445,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.13654790818691254,
      "learning_rate": 0.0001,
      "loss": 0.1157,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9346666666666666,
      "eval_loss": 0.2186373770236969,
      "eval_runtime": 6.1601,
      "eval_samples_per_second": 608.759,
      "eval_steps_per_second": 9.578,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.017136365175247192,
      "learning_rate": 0.0001,
      "loss": 0.0592,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9304,
      "eval_loss": 0.2712576687335968,
      "eval_runtime": 6.1406,
      "eval_samples_per_second": 610.691,
      "eval_steps_per_second": 9.608,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.11500805616378784,
      "learning_rate": 0.0001,
      "loss": 0.0447,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9245333333333333,
      "eval_loss": 0.315349817276001,
      "eval_runtime": 6.6516,
      "eval_samples_per_second": 563.777,
      "eval_steps_per_second": 8.87,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.02844385616481304,
      "learning_rate": 0.0001,
      "loss": 0.0352,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9389333333333333,
      "eval_loss": 0.2600131332874298,
      "eval_runtime": 6.1865,
      "eval_samples_per_second": 606.161,
      "eval_steps_per_second": 9.537,
      "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
}