trainer_state.json

{
  "best_metric": 0.9272,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0246/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "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": 9e-05,
      "loss": 0.6537,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9229333333333334,
      "eval_loss": 0.26753395795822144,
      "eval_runtime": 4.721,
      "eval_samples_per_second": 794.321,
      "eval_steps_per_second": 12.497,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 11.521432876586914,
      "learning_rate": 9e-05,
      "loss": 0.1448,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9261333333333334,
      "eval_loss": 0.26065677404403687,
      "eval_runtime": 4.791,
      "eval_samples_per_second": 782.723,
      "eval_steps_per_second": 12.315,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.08310438692569733,
      "learning_rate": 9e-05,
      "loss": 0.0862,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9224,
      "eval_loss": 0.27544546127319336,
      "eval_runtime": 4.4053,
      "eval_samples_per_second": 851.254,
      "eval_steps_per_second": 13.393,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.031766850501298904,
      "learning_rate": 9e-05,
      "loss": 0.0582,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9192,
      "eval_loss": 0.31594958901405334,
      "eval_runtime": 4.2827,
      "eval_samples_per_second": 875.617,
      "eval_steps_per_second": 13.776,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.021565886214375496,
      "learning_rate": 9e-05,
      "loss": 0.0453,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9216,
      "eval_loss": 0.3225235044956207,
      "eval_runtime": 5.2084,
      "eval_samples_per_second": 719.991,
      "eval_steps_per_second": 11.328,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.026006093248724937,
      "learning_rate": 9e-05,
      "loss": 0.0401,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9216,
      "eval_loss": 0.3434818685054779,
      "eval_runtime": 4.2853,
      "eval_samples_per_second": 875.075,
      "eval_steps_per_second": 13.768,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.2850637137889862,
      "learning_rate": 9e-05,
      "loss": 0.0357,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9272,
      "eval_loss": 0.33867257833480835,
      "eval_runtime": 4.3533,
      "eval_samples_per_second": 861.409,
      "eval_steps_per_second": 13.553,
      "step": 2331
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}