Prediction Utils

Package to house various prediction utilies.

`convert_outputs(predictions, tiled)`

Convert the Lightning trainer outputs to the desired form.

This method allows stitching back together tiled predictions.

Parameters:

Name	Type	Description	Default
`predictions`	`list`	Predictions that are output from `Trainer.predict`.	required
`tiled`	`bool`	Whether the predictions are tiled.	required

Returns:

Type	Description
`list of numpy.ndarray or numpy.ndarray`	list of arrays with the axes SC(Z)YX. If there is only 1 output it will not be in a list.

`convert_outputs_microsplit(predictions, dataset)`

Convert microsplit Lightning trainer outputs using eval_utils stitching functions.

This function processes microsplit predictions that return (tile_prediction, tile_std) tuples and stitches them back together using the same logic as get_single_file_mmse.

Parameters:

Name	Type	Description	Default
`predictions`	`list of tuple[NDArray, NDArray]`	Predictions from Lightning trainer for microsplit. Each element is a tuple of (tile_prediction, tile_std) where both are numpy arrays from predict_step.	required
`dataset`	`Dataset`	The dataset object used for prediction, needed for stitching function selection and stitching process.	required

Returns:

Type	Description
`tuple[NDArray, NDArray]`	A tuple of (stitched_predictions, stitched_stds) representing the full stitched predictions and standard deviations.

`convert_outputs_pn2v(predictions, tiled)`

Convert the Lightning trainer outputs to the desired form.

This method allows stitching back together tiled predictions.

Parameters:

Name	Type	Description	Default
`predictions`	`list`	Predictions that are output from `Trainer.predict`. Length of list the total number of tiles divided by the batch size. Each element consists of a tuple of ((prediction, mse), tile_info_list). 1st dimension of each tensor is the bs. Length of tile info list is the batch size.	required
`tiled`	`bool`	Whether the predictions are tiled.	required

Returns:

Type	Description
`tuple[list[NDArray], list[NDArray]]`	Tuple of (predictions, mmse) where each is a list of arrays with axes SC(Z)YX.

`stitch_prediction_single(tiles, tile_infos)`

Stitch tiles back together to form a full image.

Tiles are of dimensions SC(Z)YX, where C is the number of channels and can be a singleton dimension.

Parameters:

Name	Type	Description	Default
`tiles`	`list of numpy.ndarray`	Cropped tiles and their respective stitching coordinates.	required
`tile_infos`	`list of TileInformation`	List of information and coordinates obtained from `dataset.tiled_patching.extract_tiles`.	required

Returns:

Type	Description
`ndarray`	Full image, with dimensions SC(Z)YX.

`stitch_prediction_vae(predictions, dset)`

Stitch predictions back together using dataset's index manager.

Parameters:

Name	Type	Description	Default
`predictions`	`ndarray`	Array of predictions with shape (n_tiles, channels, height, width).	required
`dset`	`Dataset`	Dataset object with idx_manager containing tiling information.	required

Returns:

Type	Description
`ndarray`	Stitched predictions.