Update app.py

app.py

@@ -5,7 +5,7 @@ from sentence_transformers import SentenceTransformer
 import faiss
 import numpy as np
 
-st.title("💬 Fin$mart Chatbot")
+st.title("Fin$mart Chatbot")
 st.markdown("Ask financial questions and get answers based on expert knowledge.")
 
 # Load models
@@ -21,13 +21,13 @@ tokenizer, model, embedder = load_models()
 # Load data
 @st.cache_data
 def load_data():
-    df = pd.read_csv("train_data.csv").head(100)  # Limit to 100 rows for performance
-    df['content'] = df['answer']  # Ensure 'content' column exists
+    df = pd.read_csv("train_data.csv").head(100)  # Adjust row count if needed
+    df['content'] = df['answer']  # Ensure 'content' is mapped correctly
     return df
 
 data = load_data()
 
-# Build vector store
+# Build vector store with FAISS
 @st.cache_resource
 def build_vector_store(texts):
     embeddings = embedder.encode(texts)
@@ -43,21 +43,21 @@ index, embeddings = build_vector_store(texts)
 prompt = st.chat_input("Ask something about finance...")
 
 if prompt:
-    # Embed and search
+    # Encode the question and search for top 3 matches
     q_embed = embedder.encode([prompt])
     _, I = index.search(np.array(q_embed), k=3)
     context = " ".join([texts[i] for i in I[0]])
 
-    # Format prompt for T5
-    input_text = f"Answer the question using the context below.\nContext: {context}\nQuestion: {prompt}"
+    # Format prompt for T5 with better structure
+    input_text = f"Based on the context below, answer the question.\n\nContext: {context}\n\nQuestion: {prompt}"
     inputs = tokenizer(input_text, return_tensors="pt", truncation=True, max_length=512)
     outputs = model.generate(**inputs, max_length=100)
     answer = tokenizer.decode(outputs[0], skip_special_tokens=True)
 
-    # Display answer
+    # Display the generated answer
     st.markdown(f"**Answer:** {answer}")
 
-    # Optionally show context
-    with st.expander("🔍 Context Used"):
+    # Show retrieved content as reference
+    with st.expander(" Context Used"):
         for i in I[0]:
             st.write(texts[i])