use just one batch of pancreas data

[scottgigante-immunai]

Removes batch as a covariate, which might be causing bad performance.

Benchmark passing at https://github.com/scottgigante-immunai/openproblems/actions/runs/3322839154

[LuckyMD]

There are a lot of changes in here... and if I read correctly 3 different "techs", and not 1... no?

[codecov]

Codecov Report

Base: 94.96% // Head: 94.93% // Decreases project coverage by -0.03% ⚠️

Coverage data is based on head (fbdab9e) compared to base (235b251).
Patch coverage: 96.29% of modified lines in pull request are covered.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #649      +/-   ##
==========================================
- Coverage   94.96%   94.93%   -0.04%     
==========================================
  Files         152      152              
  Lines        3876     3867       -9     
  Branches      203      201       -2     
==========================================
- Hits         3681     3671      -10     
  Misses        128      128              
- Partials       67       68       +1     

Flag	Coverage Δ
unittests	94.93% <96.29%> (-0.04%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
openproblems/data/pancreas.py	96.66% <80.00%> (-3.34%)	⬇️
...lems/tasks/_cell_cell_communication/_common/api.py	91.83% <100.00%> (+0.25%)	⬆️
...ommunication_ligand_target/datasets/tnbc_wu2021.py	100.00% <100.00%> (ø)
...cation_source_target/datasets/allen_brain_atlas.py	100.00% <100.00%> (ø)
...s/tasks/spatial_decomposition/datasets/pancreas.py	100.00% <100.00%> (ø)

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[scottgigante-immunai]

Ping @LuckyMD I'd like to merge this tomorrow so we can run the benchmark

[LuckyMD]

Makes sense to me now... this works, but could be more flexible on keeping different batches.

[LuckyMD]

where is this summary coming from?

[scottgigante-immunai]

Defined at the top. Since it's the same (except alpha) for all three, I decided to reuse rather than redefine