trainer_state.json

{
  "best_metric": 0.8773333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0938/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": 62.4152717590332,
      "learning_rate": 0.0001,
      "loss": 2.6433,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7637333333333334,
      "eval_loss": 0.8219942450523376,
      "eval_runtime": 8.8008,
      "eval_samples_per_second": 426.096,
      "eval_steps_per_second": 6.704,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 107.57272338867188,
      "learning_rate": 0.0001,
      "loss": 0.6485,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.836,
      "eval_loss": 0.5250885486602783,
      "eval_runtime": 8.19,
      "eval_samples_per_second": 457.877,
      "eval_steps_per_second": 7.204,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 183.52088928222656,
      "learning_rate": 0.0001,
      "loss": 0.4119,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.856,
      "eval_loss": 0.4590977430343628,
      "eval_runtime": 8.4827,
      "eval_samples_per_second": 442.077,
      "eval_steps_per_second": 6.955,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 119.01852416992188,
      "learning_rate": 0.0001,
      "loss": 0.2672,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8736,
      "eval_loss": 0.4197212755680084,
      "eval_runtime": 10.4699,
      "eval_samples_per_second": 358.17,
      "eval_steps_per_second": 5.635,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 47.60611343383789,
      "learning_rate": 0.0001,
      "loss": 0.187,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8773333333333333,
      "eval_loss": 0.409157931804657,
      "eval_runtime": 8.2711,
      "eval_samples_per_second": 453.384,
      "eval_steps_per_second": 7.133,
      "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
}