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

Script to generate table of distances with confidence intervals #35

Closed AdamGleave closed 4 years ago

AdamGleave commented 4 years ago

Add CLI script that combines results from plot_epic_heatmap, plot_npec_heatmap and plot_erc_heatmap.

codecov[bot] commented 4 years ago

Codecov Report

Merging #35 into master will decrease coverage by 0.06%. The diff coverage is 82.41%.

Impacted file tree graph

@@            Coverage Diff             @@
##           master      #35      +/-   ##
==========================================
- Coverage   87.62%   87.56%   -0.07%     
==========================================
  Files          57       58       +1     
  Lines        4040     4181     +141     
==========================================
+ Hits         3540     3661     +121     
- Misses        500      520      +20
Impacted Files Coverage Δ
...ewards/analysis/dissimilarity_heatmaps/heatmaps.py 91.76% <0.00%> (-0.10%) :arrow_down:
tests/test_scripts.py 100.00% <ø> (ø)
src/evaluating_rewards/analysis/results.py 71.59% <76.47%> (+2.36%) :arrow_up:
.../analysis/dissimilarity_heatmaps/table_combined.py 83.33% <83.33%> (ø)
...ards/analysis/dissimilarity_heatmaps/cli_common.py 78.28% <90.90%> (+0.61%) :arrow_up:
...alysis/dissimilarity_heatmaps/plot_epic_heatmap.py 91.08% <100.00%> (-0.06%) :arrow_down:
...nalysis/dissimilarity_heatmaps/plot_erc_heatmap.py 89.24% <100.00%> (-0.12%) :arrow_down:
...alysis/dissimilarity_heatmaps/plot_npec_heatmap.py 89.89% <100.00%> (+0.79%) :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 93fed8c...5145f75. Read the comment docs.