trainer_state.json

{
  "best_metric": 0.8157333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0666/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": 3.3933701515197754,
      "learning_rate": 0.0005,
      "loss": 0.8273,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7986666666666666,
      "eval_loss": 0.6792948842048645,
      "eval_runtime": 12.5158,
      "eval_samples_per_second": 299.622,
      "eval_steps_per_second": 4.714,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.6245274543762207,
      "learning_rate": 0.0005,
      "loss": 0.4843,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7946666666666666,
      "eval_loss": 0.7013866901397705,
      "eval_runtime": 12.3822,
      "eval_samples_per_second": 302.855,
      "eval_steps_per_second": 4.765,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.25729331374168396,
      "learning_rate": 0.0005,
      "loss": 0.3699,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7938666666666667,
      "eval_loss": 0.7247616052627563,
      "eval_runtime": 11.6437,
      "eval_samples_per_second": 322.062,
      "eval_steps_per_second": 5.067,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0005,
      "loss": 0.3023,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7973333333333333,
      "eval_loss": 0.7708745002746582,
      "eval_runtime": 12.4091,
      "eval_samples_per_second": 302.197,
      "eval_steps_per_second": 4.755,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.17400570213794708,
      "learning_rate": 0.0005,
      "loss": 0.2595,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8157333333333333,
      "eval_loss": 0.6997114419937134,
      "eval_runtime": 12.4026,
      "eval_samples_per_second": 302.357,
      "eval_steps_per_second": 4.757,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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
}