trainer_state.json

{
  "best_metric": 0.8714666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0325/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "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": 5e-05,
      "loss": 1.9791,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7757333333333334,
      "eval_loss": 0.8978598713874817,
      "eval_runtime": 7.3833,
      "eval_samples_per_second": 507.901,
      "eval_steps_per_second": 7.991,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 14.78437328338623,
      "learning_rate": 5e-05,
      "loss": 0.6292,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8194666666666667,
      "eval_loss": 0.5963786840438843,
      "eval_runtime": 7.2537,
      "eval_samples_per_second": 516.978,
      "eval_steps_per_second": 8.134,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 9.032682418823242,
      "learning_rate": 5e-05,
      "loss": 0.3911,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8517333333333333,
      "eval_loss": 0.49261751770973206,
      "eval_runtime": 7.3126,
      "eval_samples_per_second": 512.816,
      "eval_steps_per_second": 8.068,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 5e-05,
      "loss": 0.2818,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8642666666666666,
      "eval_loss": 0.4627074897289276,
      "eval_runtime": 7.4735,
      "eval_samples_per_second": 501.772,
      "eval_steps_per_second": 7.895,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 6.48588228225708,
      "learning_rate": 5e-05,
      "loss": 0.2091,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8597333333333333,
      "eval_loss": 0.4738958775997162,
      "eval_runtime": 7.3472,
      "eval_samples_per_second": 510.398,
      "eval_steps_per_second": 8.03,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.755246639251709,
      "learning_rate": 5e-05,
      "loss": 0.1634,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8714666666666666,
      "eval_loss": 0.43426159024238586,
      "eval_runtime": 7.5012,
      "eval_samples_per_second": 499.92,
      "eval_steps_per_second": 7.865,
      "step": 1998
    }
  ],
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}