spokenlanguage / platalea

Library for training visually-grounded models of spoken language understanding.
Apache License 2.0
3 stars 1 forks source link

76 howto100m dataloader #114

Closed cwmeijer closed 3 years ago

cwmeijer commented 3 years ago

This will add howto100m data set loading functionality. For preprocessing functionality see #86. While the code of these two PRs are completely separate, their functionality is very much interdependent.

Checklist before merging:

codecov-commenter commented 3 years ago

Codecov Report

Merging #114 (b956e1d) into master (0a9f639) will decrease coverage by 1.92%. The diff coverage is 83.75%.

Impacted file tree graph

@@            Coverage Diff             @@
##           master     #114      +/-   ##
==========================================
- Coverage   72.19%   70.26%   -1.93%     
==========================================
  Files          32       39       +7     
  Lines        2251     2838     +587     
==========================================
+ Hits         1625     1994     +369     
- Misses        626      844     +218     
Impacted Files Coverage Δ
platalea/dataset.py 54.90% <73.52%> (+6.53%) :arrow_up:
platalea/utils/preprocessing.py 33.75% <90.56%> (ø)
tests/platalea/data/test_howto100mdata.py 97.82% <97.82%> (ø)
tests/platalea/utils/test_preprocessing.py 97.91% <97.91%> (ø)
platalea/experiments/config.py 94.11% <100.00%> (+0.27%) :arrow_up:
platalea/audio/preproc.py 100.00% <0.00%> (ø)
platalea/audio/melfreq.py 100.00% <0.00%> (ø)
... and 2 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 0a9f639...b956e1d. Read the comment docs.