search

HumanCompatibleAI / evaluating-rewards

Library to compare and evaluate reward functions
https://arxiv.org/abs/2006.13900
Apache License 2.0
61 stars 7 forks source link

Add script to compute distance from episode return correlation #27

Closed AdamGleave closed 4 years ago

codecov[bot] commented 4 years ago

Codecov Report

Merging #27 into master will increase coverage by 0.42%. The diff coverage is 95.86%.

Impacted file tree graph

@@            Coverage Diff             @@
##           master      #27      +/-   ##
==========================================
+ Coverage   87.23%   87.66%   +0.42%     
==========================================
  Files          54       55       +1     
  Lines        3595     3736     +141     
==========================================
+ Hits         3136     3275     +139     
- Misses        459      461       +2
Impacted Files Coverage Δ
tests/test_scripts.py 100.00% <ø> (ø)
src/evaluating_rewards/scripts/model_comparison.py 79.45% <50.00%> (ø)
...ards/analysis/dissimilarity_heatmaps/cli_common.py 67.96% <75.00%> (ø)
...ysis/dissimilarity_heatmaps/plot_return_heatmap.py 97.14% <97.14%> (ø)
...ewards/analysis/dissimilarity_heatmaps/heatmaps.py 90.58% <100.00%> (+0.58%) :arrow_up:
...lysis/dissimilarity_heatmaps/plot_canon_heatmap.py 97.10% <100.00%> (+3.84%) :arrow_up:
...alysis/dissimilarity_heatmaps/plot_epic_heatmap.py 96.29% <100.00%> (+0.04%) :arrow_up:
src/evaluating_rewards/datasets.py 100.00% <100.00%> (ø)
src/evaluating_rewards/preferences.py 99.26% <100.00%> (ø)
src/evaluating_rewards/rewards.py 95.47% <100.00%> (+0.21%) :arrow_up:
... and 7 more

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 76e1c54...242be67. Read the comment docs.