Whittington 2020

[ClementineDomine]

No description provided.

[codecov-commenter]

Codecov Report

Merging #70 (6b2fb71) into main (2234c7d) will decrease coverage by 8.05%.
The diff coverage is 7.06%.

@@            Coverage Diff             @@
##             main      #70      +/-   ##
==========================================
- Coverage   50.19%   42.14%   -8.05%     
==========================================
  Files          31       34       +3     
  Lines        2050     2515     +465     
==========================================
+ Hits         1029     1060      +31     
- Misses       1021     1455     +434     

Files Changed	Coverage Δ
neuralplayground/agents/whittington_2020.py	0.00% <0.00%> (ø)
neuralplayground/arenas/batch_environment.py	11.30% <11.30%> (ø)
neuralplayground/arenas/discritized_objects.py	12.80% <12.80%> (ø)
neuralplayground/arenas/__init__.py	100.00% <100.00%> (ø)
...euralplayground/experiments/sargolini_2006_data.py	94.04% <100.00%> (ø)

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[niksirbi]

All the automated tests and CI pass locally and on GitHub, so in the interest of proceeding with the other things we have planned, I'll approve and merge this.

However, there are two issues that arise from this PR, which I will open so we don't forget them:

• this PR adds quite a lot of lines of code, but not many tests, meaning our test coverage drops by a lot. We already have issue Increase test coverage #42 for that, but we will have to at least add some smoke tests soon after.
• related to the above: the new model depends on Pytorch, but torch is not included in the dependencies. The only reason CI didn't complain about this is that we have no tests sufficiently covering the new model. Pytorch is not an easy dependency (not just a matter of adding it to pyproject.toml). We have to be careful in specifying it, and we will probably need to update the installation instructions as well. See issue Add pytorch dependency #72