These 40 commits are when the Protocol Buffers files have changed:
| Commit: | d51279c | |
|---|---|---|
| Author: | Balkı | |
Bug Fix
The documentation is generated from this commit.
| Commit: | 6cb7bbb | |
|---|---|---|
| Author: | Balkı | |
GRPC Go Client
| Commit: | e9d0a0b | |
|---|---|---|
| Author: | Balkı | |
GRPC Go Client
| Commit: | 1a77aea | |
|---|---|---|
| Author: | Balkı | |
GRPC Language ID - Get Scores
| Commit: | f3457d8 | |
|---|---|---|
| Author: | Mehmet | |
Update proto related dependencies, Remove lite runtime from lexicon proto temporarily, apparently we need a different protoc for this now.
| Commit: | 5244100 | |
|---|---|---|
| Author: | aaa | |
Add `doNotSplitInDoubleQuotes` to preprocessing grpc service
| Commit: | cb0bd18 | |
|---|---|---|
| Author: | ahmetaa | |
pre 0.16.0 work. remove stale php stuff. refactor python grpc stuff.
| Commit: | 23db1ba | |
|---|---|---|
| Author: | aaa | |
Proto name changes
| Commit: | 1f70700 | |
|---|---|---|
| Author: | ahmetaa | |
grpc work. change simple_analysis proto to morphology proto. Add ZemberekGrpcConfiguation class. Only contains normalization related stuff for now. Remove initialization methods from normalization service.
| Commit: | 089cdbc | |
|---|---|---|
| Author: | ahmetaa | |
more grpc work. remove analysis.proto TODO: simple_analysis-> morholigical_analysis. normalization now applies noisy text normalization. later add spell checking servise separately. Not sure about the current initialization mechanism.
| Commit: | 48bcf67 | |
|---|---|---|
| Author: | aaa | |
Add "Informal" Root attribute.
| Commit: | 3d13b50 | |
|---|---|---|
| Author: | ahmetaa | |
more grpc work.
| Commit: | e258ba8 | |
|---|---|---|
| Author: | aaa | |
grpc exp continued.
| Commit: | ad7f139 | |
|---|---|---|
| Author: | aaa | |
grpc exp
| Commit: | ead746d | |
|---|---|---|
| Author: | ahmetaa | |
grpc experiment
| Commit: | 375662b | |
|---|---|---|
| Author: | ahmetaa | |
grpc experiment
| Commit: | 475392d | |
|---|---|---|
| Author: | ahmetaa | |
Add optional token boundary information to Grpc preprocessing service. Add console application for grpc server.
| Commit: | 41e441d | |
|---|---|---|
| Author: | aaa | |
Merge remote-tracking branch 'origin/master'
| Commit: | 38d8df7 | |
|---|---|---|
| Author: | aaa | |
Remove Root attribute `Special`. It is not used anymore.
| Commit: | 1db7ecb | |
|---|---|---|
| Author: | Mehmet | |
Add simple normalization grpc service.
| Commit: | 96ee5ce | |
|---|---|---|
| Author: | Mehmet | |
Add sentence extractor service. A few cosmetic fixes.
| Commit: | 49aaaec | |
|---|---|---|
| Author: | Mehmet | |
Add basic preprocessing service with tokenize method. Fix a div by zero issue in langid (needs to be checked if fix is accurate)
| Commit: | cbbbbf3 | |
|---|---|---|
| Author: | aaa | |
some changes in fasttext. change language_id proto file.
| Commit: | 60054aa | |
|---|---|---|
| Author: | Mehmet | |
Separate Language Id Service.
| Commit: | e26740f | |
|---|---|---|
| Author: | ahmetaa | |
Add simple language detection to grpc module.
| Commit: | cc81cfd | |
|---|---|---|
| Author: | aaa | |
Add person-names dictionary after cleaning up Improve disambiguation further by adding more data and shuffling training data after each iteration. Make some methods static in TurkishDictionaryLoader Change name of the morpheme from "AfterDoing" to "AfterDoingSo" to make it more compatible Remove "Plural" RootAttribute and use "ImplicitPlural". Change dictionaries and proto files accordingly. Add missing values to proto file.
| Commit: | ca8d2df | |
|---|---|---|
| Author: | mdakin | |
Add more fields to response in grpc AnalysisServer. Use strings instead of enums for now.
| Commit: | 482775d | |
|---|---|---|
| Author: | mdakin | |
Add id to morphemedata.
| Commit: | 0042b3e | |
|---|---|---|
| Author: | mdakin | |
Start adding real analysis result to grpc calls.
| Commit: | 84c558d | |
|---|---|---|
| Author: | Mehmet | |
Try to add common protos to core, unfortunately does not work yet. For some reason proto includes fails, this is a problem with maven of course.
| Commit: | 9292c9b | |
|---|---|---|
| Author: | mdakin | |
Simplify main pom, fix grpc proto
| Commit: | 7232638 | |
|---|---|---|
| Author: | mdakin | |
Remove submitted generated proto, various fixes to pom files, a skeleton grpc module.
| Commit: | 23f51a5 | |
|---|---|---|
| Author: | ahmetaa | |
Add PronunciationGuesser. It is used for guessing Turkish abbreviation pronunciations. Add PronunciationGuessed RootAttribute. Add a settable `strict` parameter to Syllable extractor.
| Commit: | 8c0d183 | |
|---|---|---|
| Author: | ahmetaa | |
sync proto. add some stuff back to TurkishAlphabet to unbreak stuff.
| Commit: | e4daa3c | |
|---|---|---|
| Author: | Ahmet | |
Add `referenceItem` mechanism to Serialization. Fix dictionary.
| Commit: | 02a6462 | |
|---|---|---|
| Author: | Mehmet | |
Synchronize proto enums with changes, add index to proto. Now can serializa / deserialize with no errors.
| Commit: | 23a9b83 | |
|---|---|---|
| Author: | Mehmet | |
Add a simple serializer - deserializer. Doesn't work 100% correctly yet. Performance is better but not great. Dictionary load time: 240ms RootLexicon creation time: 130ms.
| Commit: | be83d7d | |
|---|---|---|
| Author: | Mehmet | |
lite is ok for us (smaller generated code)
| Commit: | c939509 | |
|---|---|---|
| Author: | Mehmet | |
Fixes to proto file and maven configs.
| Commit: | 0f9289d | |
|---|---|---|
| Author: | Mehmet | |
| Committer: | Mehmet | |
Initial version of lexicon protos. Does not work yet.