trainer_state.json

{
  "best_metric": 0.8709333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0676/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": 1.7403,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8032,
      "eval_loss": 0.7468516826629639,
      "eval_runtime": 7.5897,
      "eval_samples_per_second": 494.094,
      "eval_steps_per_second": 7.774,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 7e-05,
      "loss": 0.5111,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8533333333333334,
      "eval_loss": 0.5129031538963318,
      "eval_runtime": 7.459,
      "eval_samples_per_second": 502.749,
      "eval_steps_per_second": 7.91,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 9.34404182434082,
      "learning_rate": 7e-05,
      "loss": 0.3075,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8656,
      "eval_loss": 0.4746669828891754,
      "eval_runtime": 7.3935,
      "eval_samples_per_second": 507.205,
      "eval_steps_per_second": 7.98,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.277072548866272,
      "learning_rate": 7e-05,
      "loss": 0.1941,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8656,
      "eval_loss": 0.46720919013023376,
      "eval_runtime": 7.4775,
      "eval_samples_per_second": 501.502,
      "eval_steps_per_second": 7.89,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 7.1997809410095215,
      "learning_rate": 7e-05,
      "loss": 0.1389,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8709333333333333,
      "eval_loss": 0.465941458940506,
      "eval_runtime": 7.1436,
      "eval_samples_per_second": 524.942,
      "eval_steps_per_second": 8.259,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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
}