Generic tiled TIF to SVS. Could not serialize DataArray

[tuttelikz]

• wsic version: 0.9.0
• Python version: 3.11.5
• Operating System: Windows 11

Description

Hi, thanks for the tool. I am testing Generic tiled tiff to SVS with command line. Image sample from here: https://openslide.cs.cmu.edu/download/openslide-testdata/Generic-TIFF/

What I Did

wsic convert -i CMU-1.tiff -o CMU-1-1-out.svs -c jpeg --overwrite

Output

TypeError: ('Could not serialize object of type DataArray', "<xarray.DataArray '0' (Y: 32914, X: 46000, C: 3)>\ndask.array<astype, shape=(32914, 46000, 3), dtype=uint8, chunksize=(256, 256, 3), chunktype=numpy.ndarray>\nDimensions without coordinates: Y, X, C")

Additionally

Tried:

import os
import wsic
from pathlib import Path
OPENSLIDE_PATH = r'D:\001-setup-files\openslide-win64-20231011\bin'
if hasattr(os, 'add_dll_directory'):
    # Windows
    with os.add_dll_directory(OPENSLIDE_PATH):
        import openslide
else:
    import openslide

if __name__ == '__main__':
    rootpath = Path(r"D:\wsi")
    dataset = "CMU-1-1"

    reader = wsic.readers.TIFFReader(rootpath.joinpath(dataset+".tiff"))
    writer = wsic.writers.SVSWriter(
        rootpath.joinpath(dataset+"-out.svs"),
        tile_size=(256, 256, 3),
        compression="jpeg",
        compression_level=90,
        overwrite=True,
        ome=False,
        app_mag=None,
        shape=reader.shape
    )

    writer.copy_from_reader(reader, read_tile_size=(
        256, 256), num_workers=32, timeout=30.0)

Throws same:

TypeError: ('Could not serialize object of type DataArray', "<xarray.DataArray '0' (Y: 32914, X: 46000, C: 3)>\ndask.array<astype, shape=(32914, 46000, 3), dtype=uint8, chunksize=(256, 256, 3), chunktype=numpy.ndarray>\nDimensions without coordinates: Y, X, C")

If I replace reader,

reader = wsic.readers.OpenSlideReader(rootpath.joinpath(dataset+".tiff"))

Now it throws:

RuntimeError: cannot schedule new futures after interpreter shutdown

What might be the issue, how to solve this?