trainer_state.json

{
  "best_metric": 0.8778666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0301/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": 193.92477416992188,
      "learning_rate": 7e-05,
      "loss": 3.2029,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6741333333333334,
      "eval_loss": 1.5139271020889282,
      "eval_runtime": 7.3054,
      "eval_samples_per_second": 513.321,
      "eval_steps_per_second": 8.076,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 184.9482879638672,
      "learning_rate": 7e-05,
      "loss": 0.9043,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8352,
      "eval_loss": 0.6065943241119385,
      "eval_runtime": 7.1418,
      "eval_samples_per_second": 525.08,
      "eval_steps_per_second": 8.261,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 170.8624725341797,
      "learning_rate": 7e-05,
      "loss": 0.4529,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8778666666666667,
      "eval_loss": 0.42929941415786743,
      "eval_runtime": 7.1669,
      "eval_samples_per_second": 523.236,
      "eval_steps_per_second": 8.232,
      "step": 999
    }
  ],
  "logging_steps": 500,
  "max_steps": 999,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 2.4529287718656e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}