Make IDF available as a separate step in model training

247-ai / FlashML

FlashML from [24]7.ai: A library for automated model training on Apache Spark

Apache License 2.0

1 stars 3 forks source link

Make IDF available as a separate step in model training #25

Open samikrc opened 4 years ago

samikrc commented 4 years ago

Currently IDF is available together with CountVectorizer as TF_IDF. We should have IDF available as a separate step, so as to use it with, say, HashingTF.