Add application file

app.py

@@ -0,0 +1,47 @@
+import streamlit as st
+import json
+from hogragger import Hogragger
+
+# Initialize the Hogragger class with your corpus path
+hogragger = Hogragger(corpus_path='corpus.json')  # Ensure corpus.json is in the correct path
+
+# Streamlit app title
+st.title("Hogragger Query Processor")
+
+# Input text box for the query
+query_input = st.text_area("Enter your query:", height=100)
+
+# Button to process the query
+if st.button("Process Query"):
+    if query_input:
+        st.write(f"Processing query: {query_input}")
+        
+        # Step 3: Run the query through the pipeline
+        result = hogragger.process_query(query_input)
+
+        # Step 4: Format the result as JSON
+        result_dict = {
+            "query": query_input,
+            "answer": result['answer'],
+            "question_type": result['question_type'],
+            "evidence_list": [
+                {
+                    "title": ev['title'],
+                    "fact": ev['fact'],
+                    "source": ev['source'],
+                    "url": ev['url'],
+                    "published_at": ev['published_at'],
+                    "category": ev['category']
+                }
+                for ev in result['evidence_list']  # Optional: Limit if needed
+            ]
+        }
+
+        # Convert the result to pretty JSON format
+        result_json = json.dumps(result_dict, indent=4)
+
+        # Display the JSON result on the web app
+        st.subheader("Result (JSON format):")
+        st.code(result_json, language='json')
+    else:
+        st.error("Please enter a query.")

requirements.txt

@@ -0,0 +1,7 @@
+streamlit
+sentence-transformers
+transformers
+torch
+faiss-cpu
+rank_bm25
+huggingface_hub