Update README.md

README.md

@@ -50,7 +50,8 @@ As an alternative to downloading the already precomputed model (much storage). Y
 and compute the embeddings for the dataset using:
 
 ```python
-linker = EntityMentionLinker.build("dmis-lab/biosyn-sapbert-bc5cdr-chemical", dictionary_name_or_path="ctd-chemicals", hybrid_search=True)
+from flair.models.entity_mention_linking import BioSynEntityPreprocessor
+linker = EntityMentionLinker.build("dmis-lab/biosyn-sapbert-bc5cdr-chemical", dictionary_name_or_path="ctd-chemicals", preprocessor=BioSynPreprocessor(), hybrid_search=True)
 ```
 
 This will reduce the download requirements, at the cost of computation.