| Component name | Description |
|---|---|
| str_lower | component lowercasing inputs |
| nltk_moses_tokenizer | For tokenization of texts (inputs for prediction, correct texts for training) |
| spelling_error_model | Model for generation of hypotheses |
| KenlmElector | For selection of the best hypothesis |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment