trainer_state.json

{
  "best_metric": 0.8664,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0656/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": 2.648799419403076,
      "learning_rate": 3e-05,
      "loss": 2.1302,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7677333333333334,
      "eval_loss": 1.0309293270111084,
      "eval_runtime": 6.4638,
      "eval_samples_per_second": 580.157,
      "eval_steps_per_second": 9.128,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 3e-05,
      "loss": 0.6392,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8314666666666667,
      "eval_loss": 0.6250272989273071,
      "eval_runtime": 7.1776,
      "eval_samples_per_second": 522.461,
      "eval_steps_per_second": 8.22,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 5.297699928283691,
      "learning_rate": 3e-05,
      "loss": 0.3239,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.856,
      "eval_loss": 0.5115415453910828,
      "eval_runtime": 6.4022,
      "eval_samples_per_second": 585.733,
      "eval_steps_per_second": 9.216,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.408121943473816,
      "learning_rate": 3e-05,
      "loss": 0.181,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8613333333333333,
      "eval_loss": 0.47491908073425293,
      "eval_runtime": 6.4753,
      "eval_samples_per_second": 579.123,
      "eval_steps_per_second": 9.112,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.24961203336715698,
      "learning_rate": 3e-05,
      "loss": 0.1043,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8645333333333334,
      "eval_loss": 0.47518712282180786,
      "eval_runtime": 6.9823,
      "eval_samples_per_second": 537.071,
      "eval_steps_per_second": 8.45,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.21022579073905945,
      "learning_rate": 3e-05,
      "loss": 0.0645,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8664,
      "eval_loss": 0.4918883442878723,
      "eval_runtime": 6.4575,
      "eval_samples_per_second": 580.724,
      "eval_steps_per_second": 9.137,
      "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
}