trainer_state.json

{
  "best_metric": 0.8792,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0614/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": 0.0001,
      "loss": 1.5578,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8056,
      "eval_loss": 0.7046812772750854,
      "eval_runtime": 6.3103,
      "eval_samples_per_second": 594.263,
      "eval_steps_per_second": 9.35,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 9.687017440795898,
      "learning_rate": 0.0001,
      "loss": 0.5425,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8149333333333333,
      "eval_loss": 0.6250252723693848,
      "eval_runtime": 6.3462,
      "eval_samples_per_second": 590.903,
      "eval_steps_per_second": 9.297,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0001,
      "loss": 0.3545,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.856,
      "eval_loss": 0.4990408718585968,
      "eval_runtime": 6.3081,
      "eval_samples_per_second": 594.472,
      "eval_steps_per_second": 9.353,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0001,
      "loss": 0.2514,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8645333333333334,
      "eval_loss": 0.4635707139968872,
      "eval_runtime": 6.3517,
      "eval_samples_per_second": 590.394,
      "eval_steps_per_second": 9.289,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.04317299276590347,
      "learning_rate": 0.0001,
      "loss": 0.1823,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8648,
      "eval_loss": 0.49105703830718994,
      "eval_runtime": 6.2599,
      "eval_samples_per_second": 599.05,
      "eval_steps_per_second": 9.425,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.07908706367015839,
      "learning_rate": 0.0001,
      "loss": 0.1508,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8792,
      "eval_loss": 0.4435449540615082,
      "eval_runtime": 6.8861,
      "eval_samples_per_second": 544.577,
      "eval_steps_per_second": 8.568,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 1998,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}