trainer_state.json

{
  "best_metric": 0.8805333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0665/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 123.0774154663086,
      "learning_rate": 9e-05,
      "loss": 2.9459,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6637333333333333,
      "eval_loss": 1.270351767539978,
      "eval_runtime": 7.3962,
      "eval_samples_per_second": 507.019,
      "eval_steps_per_second": 7.977,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 86.83295440673828,
      "learning_rate": 9e-05,
      "loss": 0.8686,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8112,
      "eval_loss": 0.6414398550987244,
      "eval_runtime": 7.8391,
      "eval_samples_per_second": 478.37,
      "eval_steps_per_second": 7.526,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 138.65682983398438,
      "learning_rate": 9e-05,
      "loss": 0.538,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8472,
      "eval_loss": 0.5264055132865906,
      "eval_runtime": 7.8162,
      "eval_samples_per_second": 479.771,
      "eval_steps_per_second": 7.548,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 66.83769989013672,
      "learning_rate": 9e-05,
      "loss": 0.3964,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8610666666666666,
      "eval_loss": 0.46882063150405884,
      "eval_runtime": 7.9395,
      "eval_samples_per_second": 472.322,
      "eval_steps_per_second": 7.431,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 39.67259979248047,
      "learning_rate": 9e-05,
      "loss": 0.3026,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.868,
      "eval_loss": 0.4378727972507477,
      "eval_runtime": 7.4451,
      "eval_samples_per_second": 503.687,
      "eval_steps_per_second": 7.925,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 50.348262786865234,
      "learning_rate": 9e-05,
      "loss": 0.249,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8722666666666666,
      "eval_loss": 0.4343445301055908,
      "eval_runtime": 7.9096,
      "eval_samples_per_second": 474.11,
      "eval_steps_per_second": 7.459,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 138.31192016601562,
      "learning_rate": 9e-05,
      "loss": 0.1874,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8805333333333333,
      "eval_loss": 0.43430182337760925,
      "eval_runtime": 7.9,
      "eval_samples_per_second": 474.686,
      "eval_steps_per_second": 7.468,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}