parse-evaluator fixes for issue #226

singnet / language-learning

OpenCog Unsupervised Language Learning

https://wiki.opencog.org/w/Language_learning

MIT License

31 stars 11 forks source link

parse-evaluator fixes for issue #226 #227

Closed alexei-gl closed 5 years ago

alexei-gl commented 5 years ago

New options implemented: -T strict tokenization (treat tokenization discrepancies as fatal errors) -S ignore sentences mismatch (continue evaluate even is sentences mismatch) Some bugs fixed along the way.