HumanCompatibleAI / imitation

Clean PyTorch implementations of imitation and reward learning algorithms
https://imitation.readthedocs.io/
MIT License
1.33k stars 248 forks source link

Running benchmarks #812

Closed ernestum closed 1 year ago

ernestum commented 1 year ago

Description

Add scripts to run the entire benchmark suite.

Testing

Description of how you've tested your changes.

ernestum commented 1 year ago

To give you an idea of how the output CSV and Markdown look: summary.csv summary.md

codecov[bot] commented 1 year ago

Codecov Report

Merging #812 (dcde3a9) into master (d833d9e) will increase coverage by 0.03%. The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #812      +/-   ##
==========================================
+ Coverage   95.63%   95.67%   +0.03%     
==========================================
  Files         100      102       +2     
  Lines        9582     9654      +72     
==========================================
+ Hits         9164     9236      +72     
  Misses        418      418              
Files Coverage Δ
src/imitation/data/huggingface_utils.py 98.00% <100.00%> (ø)
src/imitation/util/sacred_file_parsing.py 100.00% <100.00%> (ø)
tests/util/test_sacred_file_parsing.py 100.00% <100.00%> (ø)

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

ernestum commented 1 year ago

Great. This are the final results. Raw values as well as CSV and Markdown Summary. This is to be added as a release artifact benchmark_runs.zip !