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

Improve distance heatmap plotting and calculation #34

Closed AdamGleave closed 4 years ago

AdamGleave commented 4 years ago
codecov[bot] commented 4 years ago

Codecov Report

Merging #34 into master will increase coverage by 0.21%. The diff coverage is 88.63%.

Impacted file tree graph

@@            Coverage Diff             @@
##           master      #34      +/-   ##
==========================================
+ Coverage   87.41%   87.62%   +0.21%     
==========================================
  Files          57       57              
  Lines        4004     4040      +36     
==========================================
+ Hits         3500     3540      +40     
+ Misses        504      500       -4     
Impacted Files Coverage Δ
src/evaluating_rewards/analysis/stylesheets.py 71.42% <ø> (ø)
src/evaluating_rewards/epic_sample.py 97.84% <ø> (ø)
src/evaluating_rewards/rewards.py 96.34% <ø> (+0.24%) :arrow_up:
src/evaluating_rewards/scripts/npec_comparison.py 79.45% <ø> (ø)
...ewards/analysis/dissimilarity_heatmaps/heatmaps.py 91.86% <50.00%> (+9.10%) :arrow_up:
...s/dissimilarity_heatmaps/plot_gridworld_heatmap.py 91.07% <71.42%> (-0.68%) :arrow_down:
...ards/analysis/dissimilarity_heatmaps/cli_common.py 77.66% <78.94%> (+0.19%) :arrow_up:
...nalysis/dissimilarity_heatmaps/plot_erc_heatmap.py 89.36% <87.50%> (ø)
...alysis/dissimilarity_heatmaps/plot_npec_heatmap.py 89.10% <89.10%> (ø)
...alysis/dissimilarity_heatmaps/plot_epic_heatmap.py 91.13% <92.36%> (+1.44%) :arrow_up:
... and 8 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 667cbb4...36c9333. Read the comment docs.