Update app.py

app.py

@@ -5,7 +5,7 @@ import re
 # Load models
 emotion_classifier = pipeline("text-classification", model="bhadresh-savani/distilbert-base-uncased-emotion", top_k=1)
 intent_classifier = pipeline("zero-shot-classification", model="facebook/bart-large-mnli")
-text_generator = pipeline("text2text-generation", model="google/flan-t5-base", max_new_tokens=100)
+text_generator = pipeline("text2text-generation", model="google/flan-t5-base", max_new_tokens=150)
 
 # Candidate intents
 candidate_tasks = [
@@ -43,7 +43,11 @@ def get_content_score(text, top_intents):
     return min(score + 0.1, 1.0)
 
 def generate_reply(input_text, intent):
-    prompt = f"Generate a polite and helpful customer service reply to this intent: '{intent}'. Customer said: '{input_text}'"
+    prompt = (
+        f"Write a professional customer service response to the following customer request: '{input_text}'\n"
+        f"The request is related to: '{intent}'.\n"
+        f"Start with a polite sentence, then provide a clear solution, and end with a kind question offering further help."
+    )
     return text_generator(prompt)[0]['generated_text']
 
 # UI
@@ -60,17 +64,13 @@ if st.button("Analyze Message"):
 
             # Emotion detection
             emotion_result = emotion_classifier(user_input)
-            if isinstance(emotion_result[0], list):  # Handle top_k output format
-                emotion_data = emotion_result[0][0]
-            else:
-                emotion_data = emotion_result[0]
-
+            emotion_data = emotion_result[0][0] if isinstance(emotion_result[0], list) else emotion_result[0]
             emotion_label = emotion_data['label']
             emotion_score = get_emotion_score(emotion_label)
 
             # Intent detection
             intent_result = intent_classifier(user_input, candidate_tasks)
-            top_intents = [label for label, score in zip(intent_result['labels'], intent_result['scores']) if score > 0.2]
+            top_intents = [label for label, score in zip(intent_result['labels'], intent_result['scores']) if score > 0.15][:3]
 
             # Content score
             content_score = get_content_score(user_input, top_intents)
@@ -102,6 +102,6 @@ if st.button("Analyze Message"):
                     for intent in top_intents:
                         reply = generate_reply(user_input, intent)
                         st.markdown(f"**• {intent.capitalize()}**")
-                        st.write(f"Recommended Action: {reply}")
+                        st.write(reply)
                 else:
                     st.warning("No clear intent detected. Manual review recommended.")