nets evaluate

Computes model performance metrics using a trained keras model.

---

usage: nets evaluate [-h] [--fscaler features_scaler]
                     [--rscaler response_scaler] [--eval-window percent]
                     [--transform-chunksize count]
                     feature_data response_data model_config

positional arguments:
  feature_data          Path to the feature data myco archive (.npy).
  response_data         Path to the response data myco archive (.npy).
  model_config          Path to the myco model configuration file (.yml). From
                        the saved model directory.

optional arguments:
  -h, --help            show this help message and exit
  --fscaler features_scaler
                        Path to the scaler to apply to the feature data.
                        (default: None)
  --rscaler response_scaler
                        Path to the scaler to apply to the response data.
                        (default: None)
  --eval-window percent
                        The amount of the interior window to evaluate
                        performance on. (default: 0.5)
  --transform-chunksize count
                        Apply scaler transformations in chunks this size.
                        (default: 10000)