trainer_state.json

{
  "best_metric": 0.8882666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0054/checkpoints/checkpoint-999",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 999,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 56.8145751953125,
      "learning_rate": 0.0005,
      "loss": 1.1947,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8514666666666667,
      "eval_loss": 0.48893845081329346,
      "eval_runtime": 14.7222,
      "eval_samples_per_second": 254.717,
      "eval_steps_per_second": 4.008,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 22.74016571044922,
      "learning_rate": 0.0005,
      "loss": 0.3645,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8728,
      "eval_loss": 0.4147230088710785,
      "eval_runtime": 14.2617,
      "eval_samples_per_second": 262.942,
      "eval_steps_per_second": 4.137,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 40.966983795166016,
      "learning_rate": 0.0005,
      "loss": 0.2408,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8882666666666666,
      "eval_loss": 0.398785799741745,
      "eval_runtime": 14.3418,
      "eval_samples_per_second": 261.473,
      "eval_steps_per_second": 4.114,
      "step": 999
    }
  ],
  "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": 2.4529287718656e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}