search

SforAiDl / KD_Lib

A Pytorch Knowledge Distillation library for benchmarking and extending works in the domains of Knowledge Distillation, Pruning, and Quantization.
https://kd-lib.readthedocs.io/
MIT License
609 stars 58 forks source link

Fixes issue in DML #117

Closed NeelayS closed 3 years ago

NeelayS commented 3 years ago

Fix for #116

codecov-commenter commented 3 years ago

Codecov Report

Merging #117 (f68c252) into master (eca9657) will increase coverage by 0.34%. The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #117      +/-   ##
==========================================
+ Coverage   58.23%   58.58%   +0.34%     
==========================================
  Files          60       60              
  Lines        2112     2115       +3     
==========================================
+ Hits         1230     1239       +9     
+ Misses        882      876       -6
Impacted Files Coverage Δ
KD_Lib/KD/common/base_class.py 79.35% <100.00%> (ø)
KD_Lib/KD/vision/DML/dml.py 75.43% <100.00%> (+6.06%) :arrow_up:
NeelayS commented 3 years ago

Fails for some other algos (lottery tickets, BANN), not for DML. Can merge this as a quick fix for #116 I guess.

Het-Shah commented 3 years ago

Can you make the code coverage 100% on this now?

Het-Shah commented 3 years ago

Try removing try except from dml file -> see base.py file. Code coverage is still not 100%.