Software Open Access

MartinThoma/lidtk: Initial Release

Martin Thoma

Highlights

  • 7 classifiers available: CLD-2, Google Translate, langdetect, langid, tfidf_nn, TextCat (NLTK), char features
  • lidtk can predict, evaluate WiLI-2018
  • analyze how important unicode blocks are for different languages
Future work
  • Improve support of online classifers
  • Make switching network architectures simpler
  • Add RNN

Files (75.4 kB)
Name Size
MartinThoma/lidtk-v0.2.0.zip
md5:e24e3c9d7462c2040fa9ad0eae8f7cba
75.4 kB Download
22
3
views
downloads
All versions This version
Views 2222
Downloads 33
Data volume 226.3 kB226.3 kB
Unique views 2222
Unique downloads 22

Share

Cite as