Wolfram Neural Net Repository
Immediate Computable Access to Neural Net Models
Wolfram Sentence Splitter Net for TextSentences WL 12.0 V2
Classify if a delimiter is a sentence boundary.
Released in 2018 by Wolfram Research, this neural net is a binary classifier that takes input strings with length 15and outputs whether the delimiter in the middle (i.e. at position 8) is a sentence boundary.
Number of layers: 7 | Parameter count: 152,101 | Trained size: 688 KB |
Training Set Information
- The training dataset consisted of text from various sources with around 140k manually labeled delimiters and synthesized text with around 140k labeled delimiters.
Examples
Requirements
Wolfram Language 11.3 (March 2018) or above
Resource History
Reference
- Available from: Wolfram Research