trainer_state.json

{
  "best_metric": 0.8656,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0293/checkpoints/checkpoint-666",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 666,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 41.91993713378906,
      "learning_rate": 0.0003,
      "loss": 1.6421,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8216,
      "eval_loss": 0.6021609306335449,
      "eval_runtime": 7.4625,
      "eval_samples_per_second": 502.512,
      "eval_steps_per_second": 7.906,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 32.31134033203125,
      "learning_rate": 0.0003,
      "loss": 0.5058,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8656,
      "eval_loss": 0.45730963349342346,
      "eval_runtime": 7.4284,
      "eval_samples_per_second": 504.818,
      "eval_steps_per_second": 7.942,
      "step": 666
    }
  ],
  "logging_steps": 500,
  "max_steps": 666,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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": 1.6352858479104e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}