Class QT_ROOT_LEVEL_POOL.....is implement in both .....

[stephanielucia]

Bug description

I managed to install SLEAP on my macbook (M1 chip), I could open the GUI and label frames, but then when I ran the training it would crash at one point. So I deinstalled and reinstalled sleap s described on the webpage. However, when I try to open the GUI I get this error message:

(base) stephanie@staff-net-lfx-2982 sleap % conda activate sleap (sleap) stephanie@staff-net-lfx-2982 sleap % sleap-label objc[8335]: Class QT_ROOT_LEVEL_POOL__THESE_OBJECTS_WILL_BE_RELEASED_WHEN_QAPP_GOES_OUT_OF_SCOPE is implemented in both /Users/stephanie/miniforge3/envs/sleap/lib/libQt6Core.6.9.1.dylib (0x15dab96c8) and /Users/stephanie/miniforge3/envs/sleap/lib/libQt5Core.5.15.15.dylib (0x17a6312f8). This may cause spurious casting failures and mysterious crashes. One of the duplicates must be removed or renamed. objc[8335]: Class KeyValueObserver is implemented in both /Users/stephanie/miniforge3/envs/sleap/lib/libQt6Core.6.9.1.dylib (0x15dab96f0) and /Users/stephanie/miniforge3/envs/sleap/lib/libQt5Core.5.15.15.dylib (0x17a631320). This may cause spurious casting failures and mysterious crashes. One of the duplicates must be removed or renamed. objc[8335]: Class RunLoopModeTracker is implemented in both /Users/stephanie/miniforge3/envs/sleap/lib/libQt6Core.6.9.1.dylib (0x15dab9790) and /Users/stephanie/miniforge3/envs/sleap/lib/libQt5Core.5.15.15.dylib (0x17a631370). This may cause spurious casting failures and mysterious crashes. One of the duplicates must be removed or renamed. Saving config: /Users/stephanie/.sleap/1.4.1/preferences.yaml zsh: segmentation fault sleap-label (sleap) stephanie@staff-net-lfx-2982 sleap %

I tried deleting one of the duplicate files, but then it cannot find the files when I try to start the GUI. I tried to reinstall multple times now, even in different conda environments (with different names).

I am new to this and don't have lots of knowledge when it comes to these issues.

Help is very much appreciated!

Cheers, Steph

Expected behaviour

Actual behaviour

Your personal set up

• OS:

• Version(s):

• SLEAP installation method (listed here):
  • [ ] Conda from package
  • [ ] Conda from source
  • [ ] pip package
  • [ ] Apple Silicon Macs

Environment packages

# paste output of `pip freeze` or `conda list` here

Logs

# paste relevant logs here, if any

Screenshots

How to reproduce

1. Go to '...'
2. Click on '....'
3. Scroll down to '....'
4. See error

[eberrigan]

Hi @stephanielucia,

We don't officially support Mac M1 anymore, but from your error it looks like you have two versions of QT installed. Please delete one of them.

Best,

Elizabeth

[stephanielucia]

Hi Elizabeth

Thanks for your reply and the info. But you support newer versions of the Mac M chips? Just not the M1?

I tried that, did not work. Will try again.

Best, Stephanie

[eberrigan]

Hi @stephanielucia,

Yes the dependencies for SLEAP are pretty complicated because of the tensorflow backend which uses the GPU. We have had to maintain multiple dependency lists depending on OS and hardware. Since the Mac chips are evolving so quickly, we had to stop supporting the M1 chips to prioritize the newer chips. As of now, we are having dependency issues with M1 and M4 using our standard installation, so users will have to troubleshoot these independently. #2212

We are hoping once we update our entire backend to use pytorch the dependencies will be more manageable and we can extend support.

Best of luck and maybe try the discussions as well since others should be troubleshooting also.

Thanks,

Elizabeth