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_0277/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": 9.050623058764834e-05,
      "loss": 0.5784,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9341333333333334,
      "eval_loss": 0.22677718102931976,
      "eval_runtime": 6.3672,
      "eval_samples_per_second": 588.958,
      "eval_steps_per_second": 9.266,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6500646471977234,
      "learning_rate": 6.554054685128856e-05,
      "loss": 0.1239,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9466666666666667,
      "eval_loss": 0.18552997708320618,
      "eval_runtime": 6.321,
      "eval_samples_per_second": 593.264,
      "eval_steps_per_second": 9.334,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.03636626899242401,
      "learning_rate": 3.463890242149672e-05,
      "loss": 0.0465,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9517333333333333,
      "eval_loss": 0.1609344333410263,
      "eval_runtime": 6.7982,
      "eval_samples_per_second": 551.618,
      "eval_steps_per_second": 8.679,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.0235765278339386,
      "learning_rate": 9.604675162189414e-06,
      "loss": 0.0134,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9584,
      "eval_loss": 0.15420225262641907,
      "eval_runtime": 6.894,
      "eval_samples_per_second": 543.949,
      "eval_steps_per_second": 8.558,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.026628002524375916,
      "learning_rate": 8.900433292047261e-11,
      "loss": 0.006,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9608,
      "eval_loss": 0.1480005532503128,
      "eval_runtime": 6.287,
      "eval_samples_per_second": 596.471,
      "eval_steps_per_second": 9.384,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}