trainer_state.json

{
  "best_metric": 0.9458666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0226/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": 0.5607795715332031,
      "learning_rate": 3e-05,
      "loss": 1.1672,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9213333333333333,
      "eval_loss": 0.314700722694397,
      "eval_runtime": 6.3245,
      "eval_samples_per_second": 592.935,
      "eval_steps_per_second": 9.329,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 19.171836853027344,
      "learning_rate": 3e-05,
      "loss": 0.1717,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9410666666666667,
      "eval_loss": 0.2245205193758011,
      "eval_runtime": 6.0309,
      "eval_samples_per_second": 621.802,
      "eval_steps_per_second": 9.783,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.1476166695356369,
      "learning_rate": 3e-05,
      "loss": 0.0664,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9381333333333334,
      "eval_loss": 0.22430521249771118,
      "eval_runtime": 6.4851,
      "eval_samples_per_second": 578.252,
      "eval_steps_per_second": 9.098,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.04458606243133545,
      "learning_rate": 3e-05,
      "loss": 0.0288,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9408,
      "eval_loss": 0.2401152402162552,
      "eval_runtime": 5.837,
      "eval_samples_per_second": 642.456,
      "eval_steps_per_second": 10.108,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.056007109582424164,
      "learning_rate": 3e-05,
      "loss": 0.0137,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9458666666666666,
      "eval_loss": 0.2335721254348755,
      "eval_runtime": 6.1359,
      "eval_samples_per_second": 611.153,
      "eval_steps_per_second": 9.615,
      "step": 1665
    }
  ],
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}