Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Bug report] - iblenv install fails on opencv-python #275

Closed
anne-urai opened this issue Nov 11, 2021 · 3 comments
Closed

[Bug report] - iblenv install fails on opencv-python #275

anne-urai opened this issue Nov 11, 2021 · 3 comments
Labels
bug Something isn't working

Comments

@anne-urai
Copy link

On Mac OS, the iblenv install fails on opencv-python:
ERROR: Failed building wheel for opencv-python
Related to this issue: opencv/opencv-python#577. Looks like it may be fixed in their next release.

But for now, I fixed with commenting out that line from the .yaml file, and then pip install opencv-python==4.5.3.56 in the iblenv.

Closing this, but putting the hack here in case others may benefit.

@anne-urai anne-urai added the bug Something isn't working label Nov 11, 2021
@berkgercek
Copy link
Member

So, broadly, I've found a solution for building iblenv that just generally makes life so much easier: use miniforge exclusively, i.e. cut out all of the continuum.io default channel and rely purely on the community-supported conda-forge channel.

I used to have a ton of issues with broken dependencies surrounding several packages every time I tried to install e.g. iblenv and then pytorch, but cutting out the default channel solved everything.

Also as a bonus miniforge optionally comes prepackaged with mamba which is much, much faster than conda.

@anne-urai
Copy link
Author

@berkgercek do you have step-by-step instructions for that to put on the README? Sounds like it could save a lot of headache.

@berkgercek
Copy link
Member

berkgercek commented Nov 15, 2021

@anne-urai I can write up a set of instructions without a problem, but the broader issue is that the miniconda or mambaforge options are fairly specific to new installs. People with an existing conda installation who just want an IBL environment will not really get much mileage out of it.

@oliche @nbonacchi I can write up an "optional" install section on the README.md using the mambaforge option, with the suggestion that those who don't already have anaconda use this instead for simplicity and speed?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
None yet
Development

No branches or pull requests

2 participants