All notable changes to this project will be documented in this file.
The format is based on Keep a Changelog and this project adheres to Semantic Versioning.
- SDK component
- Model
- State
- Command line interface
- Structure
- Update component
- Update docs
- lexicon -> vsm
- POS, NER -> token_tagger
- unsed files
- Add mxnet to deps
- Inherited component from elit sdk
- Flatten structure
- Update POS tagger
- Segmenter test
- Fix Segmenter
- Seperate Segmenter out form Tokenizer
- Add filed name: tok and offset to SpaceTokenizer and EnglishTokenizer
- Update tokenizer tests
- Change result format. 'token' -> 'tok'. Add 'sid' field
- Update elitsdk to 0.0.6
- Refactor tokenizer
- Refartor tokenizer test
- Replace fasttext with yafasttext
- Deep Biaffine Attention Neural Dependency Paser by @hankcs
- NER
- NLPState and NLPModel
- Lexicon
- Add marisa_trie into dependency
- Configuration: sentiment model is not loaded by default
- Sentiment Analysis
- It is first released.