trainer_state.json

{
  "best_metric": 0.8922666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0909/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 52.071712493896484,
      "learning_rate": 0.0003,
      "loss": 1.4643,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8525333333333334,
      "eval_loss": 0.48426252603530884,
      "eval_runtime": 7.9685,
      "eval_samples_per_second": 470.605,
      "eval_steps_per_second": 7.404,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 47.85734558105469,
      "learning_rate": 0.0003,
      "loss": 0.3923,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8864,
      "eval_loss": 0.38324517011642456,
      "eval_runtime": 7.2571,
      "eval_samples_per_second": 516.734,
      "eval_steps_per_second": 8.13,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 34.34416198730469,
      "learning_rate": 0.0003,
      "loss": 0.2486,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8762666666666666,
      "eval_loss": 0.42422622442245483,
      "eval_runtime": 7.2278,
      "eval_samples_per_second": 518.83,
      "eval_steps_per_second": 8.163,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 34.31258010864258,
      "learning_rate": 0.0003,
      "loss": 0.1664,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.892,
      "eval_loss": 0.41764506697654724,
      "eval_runtime": 7.3672,
      "eval_samples_per_second": 509.012,
      "eval_steps_per_second": 8.008,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 48.12087631225586,
      "learning_rate": 0.0003,
      "loss": 0.1258,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8922666666666667,
      "eval_loss": 0.422066867351532,
      "eval_runtime": 7.2441,
      "eval_samples_per_second": 517.665,
      "eval_steps_per_second": 8.145,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}