trainer_state.json

{
  "best_metric": 0.9141333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0694/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": 140.48890686035156,
      "learning_rate": 0.0001,
      "loss": 2.6443,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8181333333333334,
      "eval_loss": 0.7337123155593872,
      "eval_runtime": 14.3479,
      "eval_samples_per_second": 261.363,
      "eval_steps_per_second": 4.112,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 100.06768035888672,
      "learning_rate": 0.0001,
      "loss": 0.5431,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8792,
      "eval_loss": 0.41780441999435425,
      "eval_runtime": 14.4913,
      "eval_samples_per_second": 258.776,
      "eval_steps_per_second": 4.071,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 64.3356704711914,
      "learning_rate": 0.0001,
      "loss": 0.325,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8989333333333334,
      "eval_loss": 0.34408098459243774,
      "eval_runtime": 14.552,
      "eval_samples_per_second": 257.697,
      "eval_steps_per_second": 4.054,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 146.96437072753906,
      "learning_rate": 0.0001,
      "loss": 0.2294,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9,
      "eval_loss": 0.3367672562599182,
      "eval_runtime": 14.4891,
      "eval_samples_per_second": 258.815,
      "eval_steps_per_second": 4.072,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 79.65142059326172,
      "learning_rate": 0.0001,
      "loss": 0.1606,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9098666666666667,
      "eval_loss": 0.3105486333370209,
      "eval_runtime": 14.4814,
      "eval_samples_per_second": 258.954,
      "eval_steps_per_second": 4.074,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 125.71892547607422,
      "learning_rate": 0.0001,
      "loss": 0.1233,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9077333333333333,
      "eval_loss": 0.32635462284088135,
      "eval_runtime": 14.3959,
      "eval_samples_per_second": 260.491,
      "eval_steps_per_second": 4.098,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 75.36900329589844,
      "learning_rate": 0.0001,
      "loss": 0.0925,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9141333333333334,
      "eval_loss": 0.32104775309562683,
      "eval_runtime": 14.412,
      "eval_samples_per_second": 260.201,
      "eval_steps_per_second": 4.094,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}