Skip to content

Troubleshooting

Encountered errors and solutions:

Module versions

  • skimage.morphology.selem module not found error => Problem of compatibility between Big-fish and scikit-image versions. Ideally, chooses a recent version of skimage (scikit-image==0.19.3) and Big-fish (big-fish==0.6.2).
  • numpy.core.multiarray failed to import (on python 3.11, when calling stardist) => Problem of compatibility between Numpy version and Stardist. You can downgrade Numpy to 1.26, Tensorflow to 2.14.
  • DNN library is not found. [[{{node model/down_level_0_no_0/Relu}}]] [Op:__inference_predict_function_14500] (on python 3.10, when calling stardist, epyseg or SepaNet) => Problem of Tensorflow version compatibility with your cuda drivers installation. See here for the compatible versions cuda-tensorflow. See also this discussion about this problem.

Weird point selection

We encountered a weird point selection on a version of napari, where the selected points did not correspond at all to the drawn selection rectangle. This is due to napari version 0.4.19, we strongly recommend to avoid this version.

Acces violation reading

OSError: exception: access violation reading 0x0000000000000034.

This error happened only in Windows with specific nvidia card (A6000). It happens when adding or deleting Shape layers, quite often in cytoplasmic measure option. It seems to be an error external to the plugin or napari. We haven't found a solution yet, but please refer to this discussion on imagesc forum for more info/updates.

Other issues

You can also check on the issues page of the repository to see if your problem has already been reported and has a solution. Otherwise, open a new one in this page and we will do our best to answer fast.

Tested and working configurations

Here we proposed the list of versions that worked fine for us. For some versions, we provide the full list of package versions that were installed on several python environments, with the corresponding operating system, that worked fine for us.

For each set-up, we list first the graphical info that we get with napari --info, then the link to the full yaml file.

Note that Epyseg cannot be install on python versions above 3.10. Thus, to use the full pipeline with all options, we recommend python 3.10. However, if you don't intend to use Epyseg or use it separately, the pipeline and the other dependencies are compatible with more recent python versions.

Environment lists

Windows 10, python 3.9.21, napari 0.4.18 

fishfeats: 1.1.11               
napari: 0.4.18
Platform: Windows-10-10.0.19045-SP0
Python: 3.9.21 | packaged by conda-forge | [MSC v.1929 64 bit (AMD64)]
Qt: 5.15.2
PyQt5: 5.15.11
NumPy: 1.26.4
SciPy: 1.13.1
Dask: 2024.8.0
VisPy: 0.12.2
magicgui: 0.9.1
superqt: 0.6.7
in-n-out: 0.2.1
app-model: 0.2.8
npe2: 0.7.7

OpenGL:
     - GL version:  4.6.0 NVIDIA 571.59
     - MAX_TEXTURE_SIZE: 32768

yaml file with all python packages installed in the environment here

Windows 10, python 3.10.18, napari 0.6.1 

napari: 0.6.1
Platform: Windows-10-10.0.19045-SP0
Python: 3.10.18 | packaged by conda-forge | MSC v.1943 64 bit (AMD64)
Qt: 5.15.2
PyQt5: 5.15.11
NumPy: 1.26.4
SciPy: 1.15.3
Dask: 2025.5.1
VisPy: 0.15.2
magicgui: 0.10.1
superqt: 0.7.5
in-n-out: 0.2.1
app-model: 0.3.2
psygnal: 0.13.0
npe2: 0.7.8
pydantic: 2.11.7

OpenGL:
     - PyOpenGL: 3.1.9
     - GL version:  4.6.0 NVIDIA 571.59
     - MAX_TEXTURE_SIZE: 32768
     - GL_MAX_3D_TEXTURE_SIZE: 16384

Optional:
      - numba: 0.61.2
      - triangle: 20250106
      - napari-plugin-manager: 0.1.6
      - bermuda: 0.1.4
      - PartSegCore not installed

Experimental Settings:
      - Async: False
      - Autoswap buffers: False
      - Triangulation backend: Fastest available

fishfeats: 1.1.11

yaml file with all python packages installed in the environment here

MacBook pro M1, python 3.10.14, napari 0.4.19 

napari: 0.4.19
Platform: macOS-15.5-arm64-arm-64bit
System: MacOS 15.5
Python: 3.10.14 | packaged by conda-forge | (main, Mar 20 2024, 12:51:49) [Clang 16.0.6 ]
Qt: 5.15.8
PyQt5: 5.15.9
NumPy: 1.26.4
SciPy: 1.13.1
Dask: 2025.5.1
VisPy: 0.14.3
magicgui: 0.10.1
superqt: 0.7.5
in-n-out: 0.2.1
app-model: 0.2.8
npe2: 0.7.8

OpenGL:
GL version:    2.1 Metal - 89.4
MAX_TEXTURE_SIZE: 16384

fishfeats: 1.1.11

yaml file with all python packages installed in the environment here

MacBook pro M1, python 3.10.14, napari 0.5.5

MacBook pro M1, python 3.10.14, napari 0.6.2

Ubuntu 20.04.6, python 3.10.0, napari 0.6.1 

napari: 0.6.1
Platform: Linux-5.15.0-139-generic-x86_64-with-glibc2.31
System: Ubuntu 20.04.6 LTS
Python: 3.10.0 | packaged by conda-forge | (default, Nov 20 2021, 02:24:10) [GCC 9.4.0]
Qt: 5.15.2
PySide2: 5.15.2.1
NumPy: 1.24.2
SciPy: 1.15.3
Dask: 2025.5.1
VisPy: 0.15.2
magicgui: 0.10.0
superqt: 0.7.3
in-n-out: 0.2.1
app-model: 0.3.1
psygnal: 0.13.0
npe2: 0.7.8
pydantic: 2.11.5

OpenGL:
    - PyOpenGL: 3.1.9
    - GL version:  4.6.0 NVIDIA 545.29.06
    - MAX_TEXTURE_SIZE: 32768
    - GL_MAX_3D_TEXTURE_SIZE: 16384

Optional:
- numba: 0.61.2
- triangle not installed
- napari-plugin-manager not installed
- bermuda not installed
- PartSegCore not installed

Experimental Settings:
- Async: False
- Autoswap buffers: False
- Triangulation backend: Fastest available

fishfeats: 1.1.3

yaml file with all python packages installed in the environment here

Ubuntu 20.04.6, python 3.11, napari 0.6.2 - No EPYSEG 

napari: 0.6.2
Platform: Linux-5.15.0-139-generic-x86_64-with-glibc2.31
System: Ubuntu 20.04.6 LTS
Python: 3.11.13 | packaged by conda-forge | (main, Jun  4 2025, 14:48:23) [GCC 13.3.0]
Qt: 5.15.14
PyQt5: 5.15.11
NumPy: 1.26.0
SciPy: 1.16.0
Dask: 2025.7.0
VisPy: 0.15.2
magicgui: 0.10.1
superqt: 0.7.5
in-n-out: 0.2.1
app-model: 0.4.0
psygnal: 0.14.0
npe2: 0.7.9
pydantic: 2.11.7

OpenGL:
    - PyOpenGL: 3.1.9
    - GL version:  4.6.0 NVIDIA 545.29.06
    - MAX_TEXTURE_SIZE: 32768
    - GL_MAX_3D_TEXTURE_SIZE: 16384

Optional:
 - numba: 0.61.2
 - triangle not installed
 - napari-plugin-manager not installed
 - bermuda not installed
 - PartSegCore not installed

Experimental Settings:
  - Async: False
  - Autoswap buffers: False
  - Triangulation backend: Fastest available

fishfeats: 1.1.15

yaml file with all python packages installed in the environment here

Ubuntu 20.04.6, python 3.9, napari 0.4.19 

napari: 0.4.19
Platform: Linux-5.15.0-139-generic-x86_64-with-glibc2.31
System: Ubuntu 20.04.6 LTS
Python: 3.9.19 | packaged by conda-forge | (main, Mar 20 2024, 12:50:21) [GCC 12.3.0]
Qt: 5.15.2
PySide2: 5.15.2.1
NumPy: 1.24.3
SciPy: 1.13.0
Dask: 2024.8.0
VisPy: 0.14.3
magicgui: 0.10.0
superqt: 0.7.3
in-n-out: 0.2.1
app-model: 0.2.8
npe2: 0.7.8

OpenGL:
- GL version: 4.6.0 NVIDIA 545.29.06
- MAX_TEXTURE_SIZE: 32768

fishfeats: 1.1.31

yaml file with all python packages installed in the environment here