find_objects throws AttributeError
#355
Comments
|
Thanks for reporting this @tibuch 🙏 Thanks to your detailed report I could reproduce: !pip install pyarrow
import dask.array as da
from dask_image import ndmeasure
bbox_df = ndmeasure.find_objects(da.ones((10, 10), chunks=(5,5), dtype='int')).compute()Turns out we had come across this issue before and weren't sure whether it's still present (as it currently escapes our CI tests, for more details see here). It's related to a recent change in dask and occurs in the presence of Until the problem is fixed, there's the following workaround: with dask.config.set({"dataframe.convert-string": False}):
bbox_df = ndmeasure.find_objects(da.ones((10, 10), chunks=(5,5), dtype='int')).compute() |
|
Am no longer able to reproduce this issue. Are others still able to?
|
|
Works for me with the latest versions 👍 |
|
Thanks Tim-Oliver! 🙏 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Describe the issue:
I am trying to use
dask_image.ndmeasure.find_objectsfor bounding box detection. I have a label-image (dtype int32) with values [0, 149] ofshape = (249, 1000, 1250).This is my dataset: labeling.zarr.zip
This is the error:
Minimal Complete Verifiable Example:
Anything else we need to know?:
Environment:
Full Env:
The text was updated successfully, but these errors were encountered: