jbkinney / mavenn

MAVE-NN: genotype-phenotype maps from multiplex assays of variant effect
MIT License
24 stars 5 forks source link

pip install doesn't work python 3.11 #31

Closed sofeasible closed 1 year ago

sofeasible commented 1 year ago

Hi, I'm working with Evan and David and having an issue where the pip install isn't working with python 3.11.3 (see below). Thank you!

% python --version Python 3.11.3 % pip install mavenn --upgrade Collecting mavenn Using cached mavenn-1.0.2-py2.py3-none-any.whl (15.4 MB) Collecting numpy<=1.23.5 (from mavenn) Downloading numpy-1.23.5-cp311-cp311-macosx_11_0_arm64.whl (13.3 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 13.3/13.3 MB 13.9 MB/s eta 0:00:00 Collecting matplotlib>=3.2.0 (from mavenn) Downloading matplotlib-3.7.1-cp311-cp311-macosx_11_0_arm64.whl (7.3 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7.3/7.3 MB 17.1 MB/s eta 0:00:00 Collecting pandas>=1.3.0 (from mavenn) Downloading pandas-2.0.2-cp311-cp311-macosx_11_0_arm64.whl (10.7 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 10.7/10.7 MB 21.5 MB/s eta 0:00:00 INFO: pip is looking at multiple versions of mavenn to determine which version is compatible with other requirements. This could take a while. Collecting mavenn Using cached mavenn-1.0.1-py2.py3-none-any.whl (15.4 MB) Collecting numpy (from mavenn) Downloading numpy-1.24.3-cp311-cp311-macosx_11_0_arm64.whl (13.8 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 13.8/13.8 MB 22.8 MB/s eta 0:00:00 Collecting mavenn Using cached mavenn-1.0.0-py2.py3-none-any.whl (15.4 MB) Using cached mavenn-0.99-py2.py3-none-any.whl (15.4 MB) Using cached mavenn-0.25-py2.py3-none-any.whl (51.1 MB) Using cached mavenn-0.24-py2.py3-none-any.whl (41.7 MB) Using cached mavenn-0.23-py2.py3-none-any.whl (41.1 MB) Using cached mavenn-0.22-py2.py3-none-any.whl (32.3 MB) INFO: pip is looking at multiple versions of mavenn to determine which version is compatible with other requirements. This could take a while. Using cached mavenn-0.21-py2.py3-none-any.whl (30.5 MB) Collecting tensorflow (from mavenn) Downloading tensorflow-2.13.0rc1-cp311-cp311-macosx_12_0_arm64.whl (2.0 kB) Collecting sklearn (from mavenn) Using cached sklearn-0.0.post5.tar.gz (3.7 kB) Preparing metadata (setup.py) ... error error: subprocess-exited-with-error × python setup.py egg_info did not run successfully. │ exit code: 1 ╰─> [18 lines of output] The 'sklearn' PyPI package is deprecated, use 'scikit-learn' rather than 'sklearn' for pip commands. Here is how to fix this error in the main use cases:

atareen commented 1 year ago

Hello, thanks.

I'll try to reproduce this issue with Python 3.11 shortly. Did you try this install (with the --upgrade) in your base conda environment or (or a site wide install on your local machine) or did you try this in a new conda environment or another virtual env? Can you also please paste a list of your installed packages by running pip list?

Thanks, Ammar

sofeasible commented 1 year ago

Hi Ammar,

I tried in a new conda environment made for mavenn. I actually just reinstalled anaconda because of python interpreter errors, so installed packages are:

Package Version


pip 23.1.2 setuptools 67.8.0 wheel 0.38.4

Cheers, Sophie

atareen commented 1 year ago

Hello, thanks. I meant what packages got installed after you ran pip install mavenn in the new conda env.

sofeasible commented 1 year ago

Yep, that's all that's in that environment (plus base).

atareen commented 1 year ago

Logging note: I met briefly with Sophie and got mavenn working on her machine. mavenn demo and tests are working/passing.

Note that in a new conda environment the following two commands need to be run: pip install mavenn pip install mavenn --upgrade

I will update this in the online documentation shortly.