Update app.py

app.py

@@ -10,6 +10,15 @@ def load_model():
   return model
   
 model = load_model()
+
+def predict_fn(text, model):
+    kw_extractor = yake.KeywordExtractor(top=50)
+    candidates = kw_extractor.extract_keywords(text)
+    keyphrases = [candidate[0] for candidate in candidates]
+    kw_model = KeyBERT(model=kw_model)
+    keywords=kw_model.extract_keywords(text, keyphrases, keyphrase_ngram_range=(1, 3), top_n=50)
+    return keywords
+    
 st.title("Patent Text Extractor")
 placeholder = st.empty()
 text_input = placeholder.text_area("Paste or write text", height=300)
@@ -34,14 +43,16 @@ button = st.button("Extract Keywords")
         #diversity = st.sidebar.slider("diversity", 0.1, 1.0, 0.6, 0.01)
         #params["use_mmr"] = True
         #params["diversity"] = diversity
-kw_extractor = yake.KeywordExtractor(top=50)
-candidates = kw_extractor.extract_keywords(text_input)
-keyphrases = [candidate[0] for candidate in candidates]
+#kw_extractor = yake.KeywordExtractor(top=50)
+#candidates = kw_extractor.extract_keywords(text_input)
+#keyphrases = [candidate[0] for candidate in candidates]
 #kw_model = KeyBERT(model="google/bigbird-pegasus-large-bigpatent")
 
+predict_fn(text, kw_model)
+
 #if keywords != []:
-keywords = model.extract_keywords(text_input,keyphrases, keyphrase_ngram_range=(1, 3), 
-                       top_n=50,stop_words='english',vectorizer=KeyphraseCountVectorizer())
+#keywords = model.extract_keywords(text_input,keyphrases, keyphrase_ngram_range=(1, 3), 
+                       #top_n=50,stop_words='english',vectorizer=KeyphraseCountVectorizer())
 if keywords != []:
     st.info("Extracted keywords")
     keywords = pd.DataFrame(keywords, columns=["Keyword", "Score"])