trainer_state.json

{
  "best_metric": 0.8586666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0768/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": 7e-05,
      "loss": 2.5129,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.688,
      "eval_loss": 1.1459729671478271,
      "eval_runtime": 6.6715,
      "eval_samples_per_second": 562.093,
      "eval_steps_per_second": 8.844,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 7e-05,
      "loss": 0.7108,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8154666666666667,
      "eval_loss": 0.6087062358856201,
      "eval_runtime": 6.2805,
      "eval_samples_per_second": 597.085,
      "eval_steps_per_second": 9.394,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 53.71958541870117,
      "learning_rate": 7e-05,
      "loss": 0.3206,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8384,
      "eval_loss": 0.5205715894699097,
      "eval_runtime": 6.2809,
      "eval_samples_per_second": 597.044,
      "eval_steps_per_second": 9.393,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.0298184156417847,
      "learning_rate": 7e-05,
      "loss": 0.1681,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8493333333333334,
      "eval_loss": 0.5167450308799744,
      "eval_runtime": 6.2806,
      "eval_samples_per_second": 597.079,
      "eval_steps_per_second": 9.394,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 10.047686576843262,
      "learning_rate": 7e-05,
      "loss": 0.0988,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8586666666666667,
      "eval_loss": 0.5013293623924255,
      "eval_runtime": 6.3965,
      "eval_samples_per_second": 586.256,
      "eval_steps_per_second": 9.224,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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
}