trainer_state.json

{
  "best_metric": 0.4962666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0101/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": 0.0005,
      "loss": 3.224,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.23733333333333334,
      "eval_loss": 2.8159613609313965,
      "eval_runtime": 6.102,
      "eval_samples_per_second": 614.553,
      "eval_steps_per_second": 9.669,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0005,
      "loss": 2.6074,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3293333333333333,
      "eval_loss": 2.4306271076202393,
      "eval_runtime": 6.0904,
      "eval_samples_per_second": 615.727,
      "eval_steps_per_second": 9.687,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 14.655644416809082,
      "learning_rate": 0.0005,
      "loss": 2.2396,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3525333333333333,
      "eval_loss": 2.3135526180267334,
      "eval_runtime": 6.204,
      "eval_samples_per_second": 604.453,
      "eval_steps_per_second": 9.51,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.875509738922119,
      "learning_rate": 0.0005,
      "loss": 1.9681,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.43093333333333333,
      "eval_loss": 2.024954080581665,
      "eval_runtime": 6.1701,
      "eval_samples_per_second": 607.766,
      "eval_steps_per_second": 9.562,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 6.706805229187012,
      "learning_rate": 0.0005,
      "loss": 1.7482,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.452,
      "eval_loss": 1.942962408065796,
      "eval_runtime": 6.1282,
      "eval_samples_per_second": 611.928,
      "eval_steps_per_second": 9.628,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0005,
      "loss": 1.5522,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.4664,
      "eval_loss": 1.8749579191207886,
      "eval_runtime": 6.1422,
      "eval_samples_per_second": 610.529,
      "eval_steps_per_second": 9.606,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 14.878764152526855,
      "learning_rate": 0.0005,
      "loss": 1.3496,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.4962666666666667,
      "eval_loss": 1.810950517654419,
      "eval_runtime": 6.5095,
      "eval_samples_per_second": 576.078,
      "eval_steps_per_second": 9.064,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}