trainer_state.json

{
  "best_metric": 0.8922666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0281/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "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": 8.730227815298347e-05,
      "loss": 1.4785,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8050666666666667,
      "eval_loss": 0.6860918998718262,
      "eval_runtime": 10.0303,
      "eval_samples_per_second": 373.867,
      "eval_steps_per_second": 5.882,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 13.641613960266113,
      "learning_rate": 7.95023019690139e-05,
      "loss": 0.4918,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8253333333333334,
      "eval_loss": 0.5930524468421936,
      "eval_runtime": 10.2123,
      "eval_samples_per_second": 367.203,
      "eval_steps_per_second": 5.777,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.35221824049949646,
      "learning_rate": 6.75408389678355e-05,
      "loss": 0.2599,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8570666666666666,
      "eval_loss": 0.5005998611450195,
      "eval_runtime": 11.0889,
      "eval_samples_per_second": 338.175,
      "eval_steps_per_second": 5.321,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.3109578788280487,
      "learning_rate": 5.286061811978304e-05,
      "loss": 0.1224,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8738666666666667,
      "eval_loss": 0.45245474576950073,
      "eval_runtime": 11.8004,
      "eval_samples_per_second": 317.785,
      "eval_steps_per_second": 5.0,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.0445244237780571,
      "learning_rate": 3.723229071611673e-05,
      "loss": 0.0578,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8752,
      "eval_loss": 0.4667479693889618,
      "eval_runtime": 10.5569,
      "eval_samples_per_second": 355.218,
      "eval_steps_per_second": 5.589,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.18721434473991394,
      "learning_rate": 2.2540863691266368e-05,
      "loss": 0.0172,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8901333333333333,
      "eval_loss": 0.4384423494338989,
      "eval_runtime": 10.2169,
      "eval_samples_per_second": 367.038,
      "eval_steps_per_second": 5.775,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.02152290754020214,
      "learning_rate": 1.055833996679956e-05,
      "loss": 0.0067,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8904,
      "eval_loss": 0.44146132469177246,
      "eval_runtime": 10.1838,
      "eval_samples_per_second": 368.233,
      "eval_steps_per_second": 5.794,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.13902661204338074,
      "learning_rate": 2.7299887471028354e-06,
      "loss": 0.0032,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8922666666666667,
      "eval_loss": 0.43280234932899475,
      "eval_runtime": 10.3891,
      "eval_samples_per_second": 360.954,
      "eval_steps_per_second": 5.679,
      "step": 2664
    }
  ],
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}