trainer_state.json

{
  "best_metric": 0.9224,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0615/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": 7.214684009552002,
      "learning_rate": 0.000288581929876693,
      "loss": 0.703,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8616,
      "eval_loss": 0.4582459628582001,
      "eval_runtime": 11.2461,
      "eval_samples_per_second": 333.449,
      "eval_steps_per_second": 5.246,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 5.41986083984375,
      "learning_rate": 0.00025606601717798207,
      "loss": 0.323,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8701333333333333,
      "eval_loss": 0.45564746856689453,
      "eval_runtime": 11.0498,
      "eval_samples_per_second": 339.372,
      "eval_steps_per_second": 5.339,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.042630597949028015,
      "learning_rate": 0.00020740251485476345,
      "loss": 0.1992,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8730666666666667,
      "eval_loss": 0.4776695966720581,
      "eval_runtime": 11.0744,
      "eval_samples_per_second": 338.618,
      "eval_steps_per_second": 5.328,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.114189147949219,
      "learning_rate": 0.00015,
      "loss": 0.1093,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8922666666666667,
      "eval_loss": 0.4118160009384155,
      "eval_runtime": 10.9827,
      "eval_samples_per_second": 341.446,
      "eval_steps_per_second": 5.372,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.16017882525920868,
      "learning_rate": 9.259748514523653e-05,
      "loss": 0.047,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9042666666666667,
      "eval_loss": 0.3837456703186035,
      "eval_runtime": 11.0533,
      "eval_samples_per_second": 339.264,
      "eval_steps_per_second": 5.338,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.06709029525518417,
      "learning_rate": 4.3933982822017876e-05,
      "loss": 0.0147,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9130666666666667,
      "eval_loss": 0.36055493354797363,
      "eval_runtime": 11.3868,
      "eval_samples_per_second": 329.328,
      "eval_steps_per_second": 5.181,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": NaN,
      "learning_rate": 1.1485859908123556e-05,
      "loss": 0.0057,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9224,
      "eval_loss": 0.3400695323944092,
      "eval_runtime": 11.9534,
      "eval_samples_per_second": 313.719,
      "eval_steps_per_second": 4.936,
      "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
}