trainer_state.json

{
  "best_metric": 0.8474666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0026/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": 149.5740509033203,
      "learning_rate": 5e-05,
      "loss": 3.6825,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.43146666666666667,
      "eval_loss": 3.182227611541748,
      "eval_runtime": 8.4586,
      "eval_samples_per_second": 443.335,
      "eval_steps_per_second": 6.975,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 204.5873260498047,
      "learning_rate": 5e-05,
      "loss": 1.8404,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7192,
      "eval_loss": 1.036423683166504,
      "eval_runtime": 7.6046,
      "eval_samples_per_second": 493.124,
      "eval_steps_per_second": 7.758,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 123.89966583251953,
      "learning_rate": 5e-05,
      "loss": 0.891,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7904,
      "eval_loss": 0.7302592396736145,
      "eval_runtime": 7.6378,
      "eval_samples_per_second": 490.977,
      "eval_steps_per_second": 7.725,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 139.12991333007812,
      "learning_rate": 5e-05,
      "loss": 0.656,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8186666666666667,
      "eval_loss": 0.6116653680801392,
      "eval_runtime": 7.6202,
      "eval_samples_per_second": 492.116,
      "eval_steps_per_second": 7.743,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 82.8167724609375,
      "learning_rate": 5e-05,
      "loss": 0.5379,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8282666666666667,
      "eval_loss": 0.5519145131111145,
      "eval_runtime": 7.7406,
      "eval_samples_per_second": 484.458,
      "eval_steps_per_second": 7.622,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 161.09681701660156,
      "learning_rate": 5e-05,
      "loss": 0.457,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8474666666666667,
      "eval_loss": 0.4956609904766083,
      "eval_runtime": 7.5218,
      "eval_samples_per_second": 498.554,
      "eval_steps_per_second": 7.844,
      "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
}