trainer_state.json

{
  "best_metric": 0.844,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0120/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": 82.06269073486328,
      "learning_rate": 0.0001,
      "loss": 2.7455,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7714666666666666,
      "eval_loss": 0.9239084720611572,
      "eval_runtime": 15.1776,
      "eval_samples_per_second": 247.074,
      "eval_steps_per_second": 3.887,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 86.005859375,
      "learning_rate": 0.0001,
      "loss": 0.6569,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.844,
      "eval_loss": 0.5071207880973816,
      "eval_runtime": 15.6121,
      "eval_samples_per_second": 240.198,
      "eval_steps_per_second": 3.779,
      "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
}