trainer_state.json

{
  "best_metric": 0.8874666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0657/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": 194.23939514160156,
      "learning_rate": 7e-05,
      "loss": 3.3726,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6034666666666667,
      "eval_loss": 2.3405251502990723,
      "eval_runtime": 7.8804,
      "eval_samples_per_second": 475.864,
      "eval_steps_per_second": 7.487,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 178.39425659179688,
      "learning_rate": 7e-05,
      "loss": 1.0809,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8293333333333334,
      "eval_loss": 0.5894122123718262,
      "eval_runtime": 7.8148,
      "eval_samples_per_second": 479.86,
      "eval_steps_per_second": 7.55,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 122.84661102294922,
      "learning_rate": 7e-05,
      "loss": 0.4926,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8581333333333333,
      "eval_loss": 0.44834744930267334,
      "eval_runtime": 7.8393,
      "eval_samples_per_second": 478.359,
      "eval_steps_per_second": 7.526,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 137.3687744140625,
      "learning_rate": 7e-05,
      "loss": 0.319,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8813333333333333,
      "eval_loss": 0.37595880031585693,
      "eval_runtime": 7.8742,
      "eval_samples_per_second": 476.237,
      "eval_steps_per_second": 7.493,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 108.11618041992188,
      "learning_rate": 7e-05,
      "loss": 0.2225,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8837333333333334,
      "eval_loss": 0.3801499605178833,
      "eval_runtime": 7.9306,
      "eval_samples_per_second": 472.851,
      "eval_steps_per_second": 7.44,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 162.357666015625,
      "learning_rate": 7e-05,
      "loss": 0.1537,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8874666666666666,
      "eval_loss": 0.3698868453502655,
      "eval_runtime": 7.8694,
      "eval_samples_per_second": 476.529,
      "eval_steps_per_second": 7.497,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}