Closed hirofumi0810 closed 3 years ago
Merging #332 (f5c97c2) into master (c6044b7) will increase coverage by
0.04%
. The diff coverage is61.53%
.
@@ Coverage Diff @@
## master #332 +/- ##
==========================================
+ Coverage 67.98% 68.02% +0.04%
==========================================
Files 96 96
Lines 10836 10797 -39
==========================================
- Hits 7367 7345 -22
+ Misses 3469 3452 -17
Impacted Files | Coverage Δ | |
---|---|---|
neural_sp/models/seq2seq/decoders/build.py | 78.94% <ø> (ø) |
|
...eural_sp/models/seq2seq/decoders/rnn_transducer.py | 77.01% <45.45%> (+0.03%) |
:arrow_up: |
neural_sp/models/seq2seq/decoders/las.py | 54.26% <47.05%> (+0.29%) |
:arrow_up: |
neural_sp/models/lm/rnnlm.py | 93.87% <100.00%> (+0.88%) |
:arrow_up: |
neural_sp/models/seq2seq/encoders/rnn.py | 76.63% <100.00%> (+0.56%) |
: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 c6044b7...f5c97c2. Read the comment docs.
Removed GRU layers in RNN encoder, LAS decoder, and RNN-T decoder.