trainer_state.json

{
  "best_metric": 0.9016,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0972/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": 8.006006006006006e-05,
      "loss": 1.59,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8192,
      "eval_loss": 0.6717187762260437,
      "eval_runtime": 7.3577,
      "eval_samples_per_second": 509.67,
      "eval_steps_per_second": 8.019,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5555357933044434,
      "learning_rate": 6.0060060060060066e-05,
      "loss": 0.4874,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8458666666666667,
      "eval_loss": 0.5280184149742126,
      "eval_runtime": 7.6849,
      "eval_samples_per_second": 487.967,
      "eval_steps_per_second": 7.677,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 4.012012012012012e-05,
      "loss": 0.2821,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8797333333333334,
      "eval_loss": 0.4274028241634369,
      "eval_runtime": 7.1819,
      "eval_samples_per_second": 522.146,
      "eval_steps_per_second": 8.215,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.7319732308387756,
      "learning_rate": 2.012012012012012e-05,
      "loss": 0.1565,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8858666666666667,
      "eval_loss": 0.39611196517944336,
      "eval_runtime": 7.235,
      "eval_samples_per_second": 518.312,
      "eval_steps_per_second": 8.155,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.26906073093414307,
      "learning_rate": 1.2012012012012013e-07,
      "loss": 0.0819,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9016,
      "eval_loss": 0.3488784432411194,
      "eval_runtime": 7.0996,
      "eval_samples_per_second": 528.201,
      "eval_steps_per_second": 8.31,
      "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
}