Create app.py

app.py

@@ -0,0 +1,38 @@
+import streamlit as st
+from transformers import pipeline
+import time
+
+def sentiment(summary):
+    pipe = pipeline("text-classification", model="WillWEI0103/CustomModel_finance_sentiment_analytics")
+    label = pipe(summary)[0]['label']
+    score = pipe(summary)[0]['score']
+    return label,score
+
+
+def main():
+    dicts={"bullish":'Positive📈',"bearish":'Negative📉','neutral':"Neutral😐"}
+    st.header("Summarize Your Finance News and Analyze Sentiment📰")
+    text=st.text_input('Input your Finance news(Max lenth<=3000): ',None,max_chars=3000)
+    #if text is not None:
+    if st.button('Conduct'):
+        st.text_area('Your Finance News: ',text,height=100)
+
+        #Stage 1: Text Summarization
+        with st.status("Processing Finance News Summarization...") as status:
+            text_summarize=pipeline("summarization", model="nickmuchi/fb-bart-large-finetuned-trade-the-event-finance-summarizer")
+            summary=text_summarize(text)[0]['summary_text']
+            status.update(label="Summarization Completed", state="complete", expanded=False)
+        st.text_area('Your Finance News Summary',summary,height=30)
+
+        #Stage 2: Sentiment Analytics
+        with st.status("Processing Sentiment Analytics..") as status:
+            label,score = sentiment(summary)
+            label=dicts[label]
+            status.update(label="Sentiment Analytics Completed", state="complete", expanded=False)
+        st.text('The Sentiment of the Finance News is: ')
+        st.text(label)
+        st.text('The Sentiment Score: ')
+        st.text(round(score,3))
+        
+if __name__ == "__main__":
+    main()