trainer_state.json

{
  "best_metric": 0.9608,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0889/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": 0.6010364890098572,
      "learning_rate": 4.752422169756048e-05,
      "loss": 0.8178,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9461333333333334,
      "eval_loss": 0.21607090532779694,
      "eval_runtime": 4.3822,
      "eval_samples_per_second": 855.731,
      "eval_steps_per_second": 13.463,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.12871889770030975,
      "learning_rate": 4.058724504646834e-05,
      "loss": 0.1155,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9485333333333333,
      "eval_loss": 0.20152786374092102,
      "eval_runtime": 4.3281,
      "eval_samples_per_second": 866.438,
      "eval_steps_per_second": 13.632,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.45405346155166626,
      "learning_rate": 3.056302334890786e-05,
      "loss": 0.0458,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9568,
      "eval_loss": 0.17705152928829193,
      "eval_runtime": 4.3018,
      "eval_samples_per_second": 871.726,
      "eval_steps_per_second": 13.715,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.21727846562862396,
      "learning_rate": 1.9436976651092144e-05,
      "loss": 0.0199,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9554666666666667,
      "eval_loss": 0.17688186466693878,
      "eval_runtime": 4.3023,
      "eval_samples_per_second": 871.618,
      "eval_steps_per_second": 13.713,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.26883476972579956,
      "learning_rate": 9.412754953531663e-06,
      "loss": 0.0096,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9573333333333334,
      "eval_loss": 0.1664845198392868,
      "eval_runtime": 4.3409,
      "eval_samples_per_second": 863.872,
      "eval_steps_per_second": 13.592,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.023804599419236183,
      "learning_rate": 2.475778302439524e-06,
      "loss": 0.0048,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9608,
      "eval_loss": 0.15685781836509705,
      "eval_runtime": 4.351,
      "eval_samples_per_second": 861.877,
      "eval_steps_per_second": 13.56,
      "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
}