Closed yongyi-wu closed 3 years ago
Merging #1470 (d308a1d) into master (def0d70) will increase coverage by
0.00%
. The diff coverage is100.00%
.
@@ Coverage Diff @@
## master #1470 +/- ##
=======================================
Coverage 85.86% 85.86%
=======================================
Files 52 52
Lines 6911 6912 +1
=======================================
+ Hits 5934 5935 +1
Misses 977 977
Impacted Files | Coverage Δ | |
---|---|---|
src/gluonnlp/op.py | 95.16% <100.00%> (+0.07%) |
: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 def0d70...d308a1d. Read the comment docs.
The documentation website for preview: http://gluon-nlp-staging.s3-accelerate.dualstack.amazonaws.com/PR1470/rel-embd/index.html
We can try to improve the docstring later.
Description
This version aligns the implementation of relative position encoding with that in mesh_tensorflow. It would also help program T5 model in the future.
Changes
mem_i - query_j
in lieu of previousquery_i - mem_j
cc @dmlc/gluon-nlp-team