trainer_state.json

{
  "best_metric": 0.9245333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0665/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": 10.035027503967285,
      "learning_rate": 9e-05,
      "loss": 0.6552,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9176,
      "eval_loss": 0.2895534634590149,
      "eval_runtime": 13.1563,
      "eval_samples_per_second": 285.034,
      "eval_steps_per_second": 4.485,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.08565553277730942,
      "learning_rate": 9e-05,
      "loss": 0.1611,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9213333333333333,
      "eval_loss": 0.2811264395713806,
      "eval_runtime": 13.1916,
      "eval_samples_per_second": 284.271,
      "eval_steps_per_second": 4.473,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.5587964057922363,
      "learning_rate": 9e-05,
      "loss": 0.092,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9234666666666667,
      "eval_loss": 0.29548290371894836,
      "eval_runtime": 13.0679,
      "eval_samples_per_second": 286.964,
      "eval_steps_per_second": 4.515,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.014661576598882675,
      "learning_rate": 9e-05,
      "loss": 0.0607,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9178666666666667,
      "eval_loss": 0.32820236682891846,
      "eval_runtime": 12.8569,
      "eval_samples_per_second": 291.672,
      "eval_steps_per_second": 4.589,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.009099111892282963,
      "learning_rate": 9e-05,
      "loss": 0.0474,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9176,
      "eval_loss": 0.34651660919189453,
      "eval_runtime": 12.971,
      "eval_samples_per_second": 289.106,
      "eval_steps_per_second": 4.549,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.8841427564620972,
      "learning_rate": 9e-05,
      "loss": 0.0425,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9221333333333334,
      "eval_loss": 0.3456425666809082,
      "eval_runtime": 13.088,
      "eval_samples_per_second": 286.522,
      "eval_steps_per_second": 4.508,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": NaN,
      "learning_rate": 9e-05,
      "loss": 0.0389,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9245333333333333,
      "eval_loss": 0.3470514416694641,
      "eval_runtime": 13.0183,
      "eval_samples_per_second": 288.057,
      "eval_steps_per_second": 4.532,
      "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
}