search

LabeliaLabs / distributed-learning-contributivity

Simulate collaborative ML scenarios, experiment multi-partner learning approaches and measure respective contributions of different datasets to model performance.
https://www.labelia.org
Apache License 2.0
56 stars 12 forks source link

Add DRFA/AFL learning approach #354

Closed HeytemBou closed 2 years ago

codecov-commenter commented 2 years ago

Codecov Report

Merging #354 (9ba9096) into master (9590aea) will increase coverage by 0.67%. The diff coverage is 96.40%.

Impacted file tree graph

@@            Coverage Diff             @@
##           master     #354      +/-   ##
==========================================
+ Coverage   79.64%   80.32%   +0.67%     
==========================================
  Files          15       15              
  Lines        3272     3410     +138     
==========================================
+ Hits         2606     2739     +133     
- Misses        666      671       +5
Impacted Files Coverage Δ
mplc/multi_partner_learning/__init__.py 100.00% <ø> (ø)
mplc/scenario.py 81.25% <66.66%> (-0.15%) :arrow_down:
mplc/utils.py 87.34% <93.33%> (+0.62%) :arrow_up:
mplc/multi_partner_learning/basic_mpl.py 88.53% <97.50%> (+2.85%) :arrow_up:
mplc/multi_partner_learning/utils.py 86.86% <100.00%> (ø)

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 9590aea...9ba9096. Read the comment docs.

HeytemBou commented 2 years ago

@HeytemBou I flagged something wrong with the new parameter active_partners_count, can you please fix it? Also, in the benchmark notebook, could please: a) put the results summary before all the logs of the experiments (to improve readibility) b) write a quick analysis of those results: what are your conclusions from them?

Ok I'll update the notebook