Update app.py

app.py

@@ -2,22 +2,25 @@ import gradio as gr
 from sentence_transformers import SentenceTransformer
 from sklearn.metrics.pairwise import cosine_similarity
 import numpy as np
-import os
+import tempfile
 
 # Initialize embedding model
 embedding_model = SentenceTransformer('all-MiniLM-L6-v2')
-knowledge_chunks = []
-chunk_embeddings = None
 
 def create_chatbot(role, context, info, conv_starter, file):
-    global knowledge_chunks, chunk_embeddings
     knowledge_chunks = []
     chunk_embeddings = None
     
-    # Load knowledge if file provided
     if file:
-        with open(file.name, 'r', encoding='utf-8') as f:
+        # Use tempfile to handle uploaded file
+        with tempfile.NamedTemporaryFile(delete=False) as temp:
+            temp.write(file.read())
+            temp_path = temp.name
+        
+        with open(temp_path, 'r', encoding='utf-8') as f:
             text = f.read()
+        os.unlink(temp_path)
+        
         knowledge_chunks = [chunk.strip() for chunk in text.split('\n\n') if chunk.strip()]
         
         if knowledge_chunks:
@@ -28,28 +31,35 @@ def create_chatbot(role, context, info, conv_starter, file):
     else:
         status = "⚠️ No file uploaded"
     
-    return status, role, context, info, conv_starter
+    return status, {
+        "role": role,
+        "context": context,
+        "info": info,
+        "conv_starter": conv_starter,
+        "knowledge": knowledge_chunks,
+        "embeddings": chunk_embeddings
+    }
 
-def respond(message, history, role, context, info):
+def respond(message, history, state):
     # Check for information requests
     if any(keyword in message.lower() for keyword in ["more info", "contact", "information", "email", "details"]):
-        return info
+        return state["info"]
     
     # Handle empty knowledge base
-    if not knowledge_chunks:
+    if not state.get("knowledge"):
         return "⚠️ Please upload knowledge base first"
     
     # Process query
     query_embedding = embedding_model.encode([message])
-    similarities = cosine_similarity(query_embedding, chunk_embeddings)[0]
+    similarities = cosine_similarity(query_embedding, state["embeddings"])[0]
     max_index = np.argmax(similarities)
     max_similarity = similarities[max_index]
     
     # Return best match if above threshold
     if max_similarity > 0.45:
-        return knowledge_chunks[max_index]
+        return state["knowledge"][max_index]
     
-    return "I can't help with that"
+    return f"{state['role']}\n{state['context']}\nI can't help with that specific question."
 
 with gr.Blocks(theme=gr.themes.Soft()) as app:
     gr.Markdown("# 🤖 Custom Chatbot Creator")
@@ -60,14 +70,20 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
             gr.Markdown("## Configuration Panel")
             
             with gr.Group():
-                role = gr.Textbox(label="Role", value="You are an assistant helpful for LLM prompting")
-                context = gr.Textbox(label="Context", value="You search for vector databases for information about prompt engineering and provide short answers.")
-                info = gr.Textbox(label="Contact Info", value="If someone asks where to find more information, tell them to contact Anchal Kumar Tarwey at email: anchalkumartarwey@gmail.com")
-                conv_starter = gr.Textbox(label="Conversation Starter", value="What are the three basic LLM prompting concepts?")
+                role = gr.Textbox(label="Role", 
+                                value="AI Assistant specialized in technical queries")
+                context = gr.Textbox(label="Context", 
+                                   value="Focus on providing concise, accurate answers based on the knowledge base")
+                info = gr.Textbox(label="Contact Info", 
+                                value="For more information, contact support@example.com")
+                conv_starter = gr.Textbox(label="Conversation Starter", 
+                                        value="Ask me about topics in the knowledge base")
             
             with gr.Group():
-                file = gr.File(label="Knowledge Base (.txt only)", file_types=[".txt"])
-                create_btn = gr.Button("Create Chatbot")
+                file = gr.File(label="Knowledge Base (.txt only)", 
+                             file_types=[".txt"], 
+                             type="binary")
+                create_btn = gr.Button("Create Chatbot", variant="primary")
             
             status = gr.Textbox(label="Status", interactive=False)
             
@@ -76,10 +92,14 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
 
         with gr.Column(scale=2):
             gr.Markdown("## Chat Interface")
+            state = gr.State({})
             chatbot = gr.ChatInterface(
-                lambda m, h: respond(m, h, role.value, context.value, info.value),
-                chatbot=gr.Chatbot(height=400, avatar_images=("user.png", "bot.png")),
-                textbox=gr.Textbox(placeholder="Type your message...", container=False),
+                respond,
+                chatbot=gr.Chatbot(height=500, 
+                                 avatar_images=(None, (None, "https://i.imgur.com/7kQEsHU.png"))),
+                textbox=gr.Textbox(placeholder="Type your message...", 
+                                 container=False, 
+                                 autofocus=True),
                 submit_btn="Ask",
                 retry_btn=None,
                 undo_btn=None,
@@ -89,8 +109,8 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
     create_btn.click(
         create_chatbot,
         inputs=[role, context, info, conv_starter, file],
-        outputs=[status, role, context, info, conv_starter]
+        outputs=[status, state]
     )
 
 if __name__ == "__main__":
-    app.launch()
+    app.launch()