trainer_state.json

{
  "best_metric": 0.8282666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0166/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 53.117652893066406,
      "learning_rate": 3e-05,
      "loss": 3.851,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.3664,
      "eval_loss": 3.6975932121276855,
      "eval_runtime": 9.3387,
      "eval_samples_per_second": 401.554,
      "eval_steps_per_second": 6.318,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 144.73451232910156,
      "learning_rate": 3e-05,
      "loss": 3.1121,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5578666666666666,
      "eval_loss": 2.382776975631714,
      "eval_runtime": 8.8492,
      "eval_samples_per_second": 423.766,
      "eval_steps_per_second": 6.667,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 156.17364501953125,
      "learning_rate": 3e-05,
      "loss": 1.6714,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6872,
      "eval_loss": 1.2168947458267212,
      "eval_runtime": 8.6266,
      "eval_samples_per_second": 434.7,
      "eval_steps_per_second": 6.839,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 216.19583129882812,
      "learning_rate": 3e-05,
      "loss": 1.0428,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7637333333333334,
      "eval_loss": 0.8612083196640015,
      "eval_runtime": 9.5799,
      "eval_samples_per_second": 391.443,
      "eval_steps_per_second": 6.159,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 244.38084411621094,
      "learning_rate": 3e-05,
      "loss": 0.7737,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7957333333333333,
      "eval_loss": 0.7011486887931824,
      "eval_runtime": 8.6879,
      "eval_samples_per_second": 431.634,
      "eval_steps_per_second": 6.791,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 245.93295288085938,
      "learning_rate": 3e-05,
      "loss": 0.6219,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8282666666666667,
      "eval_loss": 0.585792064666748,
      "eval_runtime": 8.7473,
      "eval_samples_per_second": 428.702,
      "eval_steps_per_second": 6.745,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 1998,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}