Update app.py

app.py

@@ -157,7 +157,7 @@ def normalize_text(text):
     words = word_tokenize(text)
     return ' '.join(words)
 
-llm = HuggingFaceEndpoint(
+chat_model = HuggingFaceEndpoint(
     repo_id="HuggingFaceH4/starchat2-15b-v0.1",
     task="text-generation",
     max_new_tokens=4096,
@@ -167,7 +167,7 @@ llm = HuggingFaceEndpoint(
     repetition_penalty=1.2,
     do_sample=True,
 )
-chat_model = ChatHuggingFace(llm=llm)
+# chat_model = ChatHuggingFace(llm=llm)
 
 model_name = "sentence-transformers/all-mpnet-base-v2"
 embedding_llm = SentenceTransformerEmbeddings(model_name=model_name)
@@ -208,13 +208,13 @@ def Chat_Message(history):
     message=HumanMessage(content=history[-1][0])
     messages1.append(message)
     response = chat_model.invoke(messages1)
-    messages1.append(AIMessage(content=response.content))
+    messages1.append(AIMessage(content=response))
 
     if len(messages1) >= 8:
       messages1 = messages1[-8:]
 
     history[-1][1] = ""
-    for character in response.content:
+    for character in response:
         history[-1][1] += character
         time.sleep(0.0025)
         yield history
@@ -243,13 +243,13 @@ def Web_Search(history):
     msg=HumanMessage(content=augmented_prompt)
     messages2.append(msg)
     response = chat_model.invoke(messages2)
-    messages2.append(AIMessage(content=response.content))
+    messages2.append(AIMessage(content=response))
 
     if len(messages2) >= 8:
       messages2 = messages2[-8:]
 
     history[-1][1] = ""
-    for character in response.content:
+    for character in response:
         history[-1][1] += character
         time.sleep(0.0025)
         yield history
@@ -271,12 +271,12 @@ def Chart_Generator(history):
             messages3.append(prompt)
     
             response = chat_model.invoke(messages3)
-            messages3.append(AIMessage(content=response.content))
+            messages3.append(AIMessage(content=response))
     
             if len(messages3) >= 8:
                 messages3 = messages3[-8:]
     
-            combined_content = f'{image_html}<br>{response.content}'
+            combined_content = f'{image_html}<br>{response}'
         else:
             response_text = "Can't generate this image. Please provide valid chart details."
             combined_content = response_text
@@ -285,12 +285,12 @@ def Chart_Generator(history):
         messages3.append(prompt)
     
         response = chat_model.invoke(messages3)
-        messages3.append(AIMessage(content=response.content))
+        messages3.append(AIMessage(content=response))
     
         if len(messages3) >= 8:
             messages3 = messages3[-8:]
 
-        combined_content=response.content
+        combined_content=response
         
     history[-1][1] = ""
     for character in combined_content:
@@ -333,12 +333,12 @@ def Link_Scratch(history):
             message = HumanMessage(content=augmented_prompt)
             messages4.append(message)
             response = chat_model.invoke(messages4)
-            messages4.append(AIMessage(content=response.content))
+            messages4.append(AIMessage(content=response))
 
             if len(messages4) >= 1:
                 messages4 = messages4[-1:]
 
-            response_message = response.content
+            response_message = response
 
     history[-1][1] = ""
     for character in response_message:
@@ -391,12 +391,12 @@ def File_Interact(history,filepath):
             message = HumanMessage(content=augmented_prompt)
             messages5.append(message)
             response = chat_model.invoke(messages5)
-            messages5.append(AIMessage(content=response.content))
+            messages5.append(AIMessage(content=response))
 
             if len(messages5) >= 1:
                 messages5 = messages5[-1:]
 
-            response_message = response.content
+            response_message = response
 
     history[-1][1] = ""
     for character in response_message:
@@ -510,4 +510,4 @@ with gr.Blocks(theme=gr.themes.Soft()) as demo:
     chatbot.like(print_like_dislike, None, None)
 
 demo.queue(max_size=10, default_concurrency_limit=4)
-demo.launch(max_file_size="5mb",show_api=False,max_threads=50)
+demo.launch(max_file_size="5mb", show_api=False, max_threads=50)