Update README.md

README.md

@@ -13,6 +13,31 @@ widget:
 
 The [NoHarm-Anony - De-Identification of Clinical Notes Using Contextualized Language Models and a Token Classifier](https://link.springer.com/chapter/10.1007/978-3-030-91699-2_3) paper contains Flair-based models for Portuguese Language, initialized with [Flair BBP](https://github.com/jneto04/ner-pt) & trained on clinical notes with names tagged. 
 
+### Demo: How to use in Flair
+
+Requires: **[Flair](https://github.com/flairNLP/flair/)** (`pip install flair`)
+
+```python
+from flair.data import Sentence
+from flair.models import SequenceTagger
+# load tagger
+tagger = SequenceTagger.load("noharm-ai/anony")
+# make example sentence
+sentence = Sentence("FISIOTERAPIA TRAUMATO - MANHÃ  Henrique Dias, 38 anos. Exercícios metabólicos de extremidades inferiores. Realizo mobilização patelar e leve mobilização de flexão de joelho conforme liberado pelo Dr Marcelo Arocha. Oriento cuidados e posicionamentos.")
+# predict NER tags
+tagger.predict(sentence)
+# print sentence
+print(sentence)
+# print predicted NER spans
+print('The following NER tags are found:')
+# iterate over entities and print
+for entity in sentence.get_spans('ner'):
+    print(entity)
+```
+
+
+
+
 ## More Information
 
 Refer to the original paper, [De-Identification of Clinical Notes Using Contextualized Language Models and a Token Classifier](https://link.springer.com/chapter/10.1007/978-3-030-91699-2_3) for additional details and performance.