Update app.py

app.py

@@ -117,24 +117,46 @@ top_results = get_top_chunks('What causes skin cancer?',chunk_embeddings, cleane
 print(top_results)# Print the top results
 
 #the og code from gen ai lesson
-client = InferenceClient("microsoft/phi-4")
+#client = InferenceClient("microsoft/phi-4")
 # name of llm chatbot accessed ^^ or can use ' microsoft/phi-4 that's connected to the microsoft phi gen model
 
-def respond(message,history):
+#def respond(message,history):
     
-    info = get_top_chunks(message, chunk_embeddings, cleaned_chunks)
-    messages = [{'role': 'system','content':f'You are a friendly chatbot using {info} to answer questions. You are always willing to help and want the best for the user. You need to emphasize that you are not a medical professional at the end of the message, but you are here to help to the best of your ability. Be confident and comforting to the users when helping them. In your response add suggestions for acouple follow up questions to further the conversation with the chatbot.'}]
+ #   info = get_top_chunks(message, chunk_embeddings, cleaned_chunks)
+  #  messages = [{'role': 'system','content':f'You are a friendly chatbot using {info} to answer questions. You are always willing to help and want the best for the user. You need to emphasize that you are not a medical professional at the end of the message, but you are here to help to the best of your ability. Be confident and comforting to the users when helping them. In your response add suggestions for acouple follow up questions to further the conversation with the chatbot.'}]
     #use string interporlation with variable info
 
+   # if history:
+    #    messages.extend(history)
+#
+ #   messages.append({'role': 'user','content': message})
+
+  #  response = client.chat_completion(messages, max_tokens = 500, top_p=0.8)
+    #max tokens is a parameter to determine how long the message should be
+
+   # return response['choices'][0]['message']['content'].strip()
+
+from huggingface_hub import InferenceClient
+
+# Create client without passing model name here
+client = InferenceClient()
+
+def respond(message, history):
+    info = get_top_chunks(message, chunk_embeddings, cleaned_chunks)
+    messages = [{'role': 'system', 'content': f'You are a friendly chatbot using {info} ...'}]
     if history:
         messages.extend(history)
+    messages.append({'role': 'user', 'content': message})
 
-    messages.append({'role': 'user','content': message})
-
-    response = client.chat_completion(messages, max_tokens = 500, top_p=0.8)
-    #max tokens is a parameter to determine how long the message should be
+    # Use the chat completions API
+    response = client.chat.completions.create(
+        model="microsoft/phi-4",
+        messages=messages,
+        max_tokens=500,
+        top_p=0.8
+    )
+    return response.choices[0].message.content.strip()
 
-    return response['choices'][0]['message']['content'].strip()
     
 with gr.Blocks(theme= theme) as chatbot:
     with gr.Row(scale = 1):