Updated pyproject.toml and conda environment yaml files

[mese79]

• I reduced fixed pinned versions as much as possible
• Environment yaml files now use only conda-forge channel:
  • pyttorch, torchvision, and segment-anything will be installed using conda-forge channels as well.
• We still need to separate cpu and gpu yaml files:
  • On our GPU linux machine (cuda 11.8), I couldn't install pytorch for the GPU using just pytorch in the yaml file.
  • Adding pytorch-gpu and cuda-version>=11.8 to the env_gpu.yml solved the issue.

Note: There is a bug in napari v0.5.6 that if you hit the close button on any napari pop-up notifications, will crash the napari. It doesn't happen on Mac OS though. There is a PR to fix this, but still it's not in released version.

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 64.70588% with 6 lines in your changes missing coverage. Please review.

Project coverage is 24.95%. Comparing base (d2a2065) to head (8e8d460).

Files with missing lines	Patch %	Lines
src/featureforest/utils/data.py	64.70%	6 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #40      +/-   ##
==========================================
+ Coverage   24.89%   24.95%   +0.06%     
==========================================
  Files          40       40              
  Lines        2386     2388       +2     
==========================================
+ Hits          594      596       +2     
  Misses       1792     1792              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[psobolewskiPhD]

Thanks for this!
Curious

On our GPU linux machine (cuda 11.8), I couldn't install pytorch for the GPU using just pytorch in the yaml file.

if you run conda info do you have a __cuda virtual package listed?
is your conda installation old or originally a miniconda/anacoda?

[mese79]

I usually keep my conda/mamba updated. I have a miniforge mabma installation.

          conda version : 25.3.1
    conda-build version : not installed
         python version : 3.10.14.final.0
                 solver : libmamba (default)
       virtual packages : __archspec=1=zen3
                          __conda=25.3.1=0
                          __cuda=11.6=0
                          __glibc=2.28=0
                          __linux=4.18.0=0
                          __unix=0=0