haesleinhuepf / stackview

Interactive image stack viewing in jupyter notebooks based on ipycanvas and ipywidgets
BSD 3-Clause "New" or "Revised" License
130 stars 10 forks source link

Error displaying widget: model not found #28

Closed bpavie closed 8 months ago

bpavie commented 1 year ago

Hi

I tried to run it on our university jupyter lab via ondemands, I installed a conda environement and stackivew via conda install -c conda-forge stackview and ran the following example

import stackview

import numpy as np
from skimage.io import imread
from skimage.filters import gaussian

image = imread('https://github.com/haesleinhuepf/stackview/blob/main/docs/data/Haase_MRT_tfl3d1.tif?raw=true', plugin='tifffile')
stackview.slice(image, continuous_update=True)

which give me the following error message (with no extra information)

Error displaying widget: model not found

Is there something else that need to be installed ? Would you have any tips to identify the issue?

Note that it works fine locally for me via vscode

haesleinhuepf commented 1 year ago

Hi @bpavie ,

we've had similar issues on our cluster: https://github.com/BiAPoL/PoL-BioImage-Analysis-TS-GPU-Accelerated-Image-Analysis/issues/5

... and couldn't resolve it. I presume it is related to a version mismatch of the Jupyter Hub and your kernel environment. If you find out anything, I'd love to learn more!

Best, Robert

haesleinhuepf commented 1 year ago

Hi Benjamin @bpavie ,

I may have found a solution. Within Jupyter Hub, open a console and install stackview using the --user command:

pip install stackview --user

Do not activate any environment before installing it. It is important to install it in the same space where Jupyter is installed.

Edit: The full instructions I needed to install it on our cluster are documented here

Let me know if this works!

Best, Robert

haesleinhuepf commented 8 months ago

I presume this has been solved.