predict_images#

from pssr.predict import predict_images
pssr.predict.predict_images(model: Module, dataset: Dataset, device: str = 'cpu', norm: bool = False, prefix: str = None, out_dir: str = 'preds', callbacks=None)#

Predicts high-resolution images from low-resolution images using a given model.

Only uses evaluation images if applicable. Set val_split=1 in dataset to use all images.

Parameters:

  • model (nn.Module) – Model to recieve low-resolution images.

  • dataset (Dataset) – Dataset to load low-resolution images from.

  • device (str) – Device to train model on. Default is “cpu”.

  • norm (bool) – Whether to normalize prediction image intensities to ground truth, which must be provided by a paired dataset. Default is False.

  • prefix (str) – Prefix to append at the beginning the output file name. Default is None.

  • out_dir (str) – Directory to save images. A value of None returns images. Default is “preds”.

  • callbacks (list[Callable]) – Callbacks after each prediction. Can optionally specify an argument for locals to be passed. Default is None.

Returns:

Returns predicted images if out_dir is None.

Return type:

images (list[np.ndarray])