adding presidio

app.py

@@ -2,6 +2,9 @@ import streamlit as st
 import streamlit as st
 from st_aggrid import AgGrid
 import pandas as pd
+#for the PII masking
+
+
 
 #resource list to display after the assessment is complete
 resourceList = pd.read_csv('resources.csv')
@@ -23,6 +26,11 @@ with open('auditorytext.csv', newline='') as csvfile:
 #set page config
 st.set_page_config(page_title='auditory skills resources', page_icon='icon-128x128.png')
 
+#masking set up
+
+# Call analyzer to get results
+
+
 #define session variables
 if 'one' not in st.session_state:
     st.session_state['one'] = 'value'
@@ -108,6 +116,9 @@ def main():
             submitted = st.form_submit_button("Get Results")
             if submitted:
                 #load everything
+                from presidio_analyzer import AnalyzerEngine
+                from presidio_anonymizer import AnonymizerEngine
+                analyzer = AnalyzerEngine()
                 from transformers import pipeline
 
                 #define classifier for zero shot classification
@@ -119,7 +130,13 @@ def main():
                 while x < 35:
                     if details[x] != "none":
                         st.markdown(questions[x])
-                        st.markdown("**"+details[x]+"**")
+                        text=details[x]
+                        results = analyzer.analyze(text=text,
+                                                   language='en')
+                        anonymizer = AnonymizerEngine()
+                        anonymized_text = anonymizer.anonymize(text=text,analyzer_results=results)
+                        maskedText = anonymized_text.text
+                        st.markdown("**"+maskedText+"**")
                         sequence = details[x]
                         print(details[x])
                         output = classifier(sequence, sequence_labels)