Update: fixing bug raised around stanza model not having certain words in vocabulary, along with efforts to improve latency - #115
Update: fixing bug raised around stanza model not having certain words in vocabulary, along with efforts to improve latency #115AchintyaX wants to merge 11 commits into
Conversation
|
Sentry link for the bug |
… replacing with fuzzy search
|
Added fuzzy search, using |
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #115 +/- ##
=========================================
Coverage 100.00% 100.00%
=========================================
Files 47 47
Lines 1921 1925 +4
=========================================
+ Hits 1921 1925 +4
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
| """ | ||
| value = value + str(word.text) + " " | ||
| if value != "": | ||
| matches = process.extract(value, entity_patterns) |
There was a problem hiding this comment.
Make it a flag to get all entities or 1 entity
AchintyaX
left a comment
There was a problem hiding this comment.
use pyinstrument for profiling the changes and speed gains
…e which was happening earlier
…and switched to extractOne method in fuzzy searching
…and switched to extractOne method in fuzzy searching, and updating type check Signed-off-by: Achintya Shankhdhar <achintyashankhdhar@gmail.com>
6534fb1 to
85fa241
Compare
…and switched to extractOne method in fuzzy searching, and updating type check, adding PatternList Signed-off-by: Achintya Shankhdhar <achintyashankhdhar@gmail.com>
85fa241 to
7bcd692
Compare
…and switched to extractOne method in fuzzy searching, and updating type check, adding PatternList
ebffdfa to
5f836e5
Compare
A bug was found when testing in Oppo when users are saying something in Hindi but English ASR is used. certain words are out of vocabulary for the stanza model.
the PR is a error handling for this instance