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 CANON experiment configuration for PointMaze transfer #24

Closed AdamGleave closed 4 years ago

AdamGleave commented 4 years ago

Add point_maze_learned configuration to plot_canon_heatmap.py to compare learned rewards to ground-truth.

codecov[bot] commented 4 years ago

Codecov Report

Merging #24 into master will decrease coverage by 0.12%. The diff coverage is 70.58%.

Impacted file tree graph

@@            Coverage Diff             @@
##           master      #24      +/-   ##
==========================================
- Coverage   87.43%   87.30%   -0.13%     
==========================================
  Files          54       54              
  Lines        3588     3600      +12     
==========================================
+ Hits         3137     3143       +6     
- Misses        451      457       +6
Impacted Files Coverage Δ
...alysis/dissimilarity_heatmaps/plot_epic_heatmap.py 96.25% <ø> (-0.10%) :arrow_down:
...ds/analysis/dissimilarity_heatmaps/reward_masks.py 92.00% <ø> (ø)
src/evaluating_rewards/analysis/results.py 69.23% <ø> (ø)
..._rewards/analysis/dissimilarity_heatmaps/config.py 66.23% <50.00%> (ø)
...lysis/dissimilarity_heatmaps/plot_canon_heatmap.py 93.25% <69.56%> (-4.05%) :arrow_down:
...ewards/analysis/dissimilarity_heatmaps/heatmaps.py 90.00% <77.77%> (+1.11%) :arrow_up:

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 ca22c21...38e93d1. Read the comment docs.