trainer_state.json

{
  "best_metric": 0.9304,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0882/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": 1.20032799243927,
      "learning_rate": 5e-05,
      "loss": 0.8745,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9176,
      "eval_loss": 0.2869449257850647,
      "eval_runtime": 11.0018,
      "eval_samples_per_second": 340.854,
      "eval_steps_per_second": 5.363,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.06142304465174675,
      "learning_rate": 5e-05,
      "loss": 0.1792,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9210666666666667,
      "eval_loss": 0.2372266799211502,
      "eval_runtime": 10.8174,
      "eval_samples_per_second": 346.663,
      "eval_steps_per_second": 5.454,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 5.2145304679870605,
      "learning_rate": 5e-05,
      "loss": 0.0967,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9149333333333334,
      "eval_loss": 0.27132153511047363,
      "eval_runtime": 11.089,
      "eval_samples_per_second": 338.173,
      "eval_steps_per_second": 5.321,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.13811883330345154,
      "learning_rate": 5e-05,
      "loss": 0.0627,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9298666666666666,
      "eval_loss": 0.2458430826663971,
      "eval_runtime": 10.9758,
      "eval_samples_per_second": 341.66,
      "eval_steps_per_second": 5.375,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.014862309210002422,
      "learning_rate": 5e-05,
      "loss": 0.0404,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9304,
      "eval_loss": 0.2685786187648773,
      "eval_runtime": 10.8362,
      "eval_samples_per_second": 346.061,
      "eval_steps_per_second": 5.445,
      "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
}