Update app.py

app.py

@@ -1,10 +1,11 @@
 import os
 import logging
-import streamlit as st
+import gradio as gr
+from huggingface_hub import hf_hub_download
 
 # Install necessary libraries using os.system
 os.system("pip install --upgrade pip")
-os.system("pip install streamlit llama-cpp-agent huggingface_hub trafilatura beautifulsoup4 requests duckduckgo-search googlesearch-python")
+os.system("pip install llama-cpp-agent huggingface_hub trafilatura beautifulsoup4 requests duckduckgo-search googlesearch-python")
 
 # Attempt to import all required modules
 try:
@@ -22,22 +23,7 @@ try:
     from utils import CitingSources
     from settings import get_context_by_model, get_messages_formatter_type
 except ImportError as e:
-    st.error(f"Error importing modules: {e}")
-    if 'utils' in str(e):
-        st.warning("Mocking utils.CitingSources")
-        class CitingSources:
-            sources = []
-
-    if 'settings' in str(e):
-        st.warning("Mocking settings functions")
-        def get_context_by_model(model):
-            return 4096
-
-        def get_messages_formatter_type(model):
-            return MessagesFormatterType.BASIC
-
-import logging
-from huggingface_hub import hf_hub_download
+    raise ImportError(f"Error importing modules: {e}")
 
 # Download the models
 hf_hub_download(
@@ -57,7 +43,7 @@ hf_hub_download(
 )
 
 # Function to respond to user messages
-def respond(message, history, temperature, top_p, top_k, repeat_penalty):
+def respond(message, temperature, top_p, top_k, repeat_penalty):
     try:
         model = "mixtral-8x7b-instruct-v0.1.Q5_K_M.gguf"
         max_tokens = 3000
@@ -106,12 +92,6 @@ def respond(message, history, temperature, top_p, top_k, repeat_penalty):
 
         messages = BasicChatHistory()
 
-        for msn in history:
-            user = {"role": Roles.user, "content": msn[0]}
-            assistant = {"role": Roles.assistant, "content": msn[1]}
-            messages.add_message(user)
-            messages.add_message(assistant)
-
         result = web_search_agent.get_chat_response(
             message,
             llm_sampling_settings=settings,
@@ -136,7 +116,6 @@ def respond(message, history, temperature, top_p, top_k, repeat_penalty):
 
         for text in response_text:
             outputs += text
-            yield outputs
 
         output_settings = LlmStructuredOutputSettings.from_pydantic_models(
             [CitingSources], LlmStructuredOutputType.object_instance
@@ -153,23 +132,24 @@ def respond(message, history, temperature, top_p, top_k, repeat_penalty):
         )
         outputs += "\n\nSources:\n"
         outputs += "\n".join(citing_sources.sources)
-        yield outputs
+        return outputs
 
     except Exception as e:
-        st.error(f"An error occurred: {e}")
-
-st.title("Novav2 Web Engine")
-
-message = st.text_input("Enter your message:")
-history = st.session_state.get("history", [])
-temperature = st.slider("Temperature", min_value=0.1, max_value=1.0, value=0.45, step=0.1)
-top_p = st.slider("Top-p", min_value=0.1, max_value=1.0, value=0.95, step=0.05)
-top_k = st.slider("Top-k", min_value=0, max_value=100, value=40, step=1)
-repeat_penalty = st.slider("Repetition penalty", min_value=0.0, max_value=2.0, value=1.1, step=0.1)
-
-if st.button("Send"):
-    response_generator = respond(message, history, temperature, top_p, top_k, repeat_penalty)
-    for response in response_generator:
-        st.write(response)
-        history.append((message, response))
-        st.session_state["history"] = history
+        return f"An error occurred: {e}"
+
+# Gradio interface
+demo = gr.Interface(
+    fn=respond,
+    inputs=[
+        gr.Textbox(label="Enter your message:"),
+        gr.Slider(minimum=0.1, maximum=1.0, value=0.45, step=0.1, label="Temperature"),
+        gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p"),
+        gr.Slider(minimum=0, maximum=100, value=40, step=1, label="Top-k"),
+        gr.Slider(minimum=0.0, maximum=2.0, value=1.1, step=0.1, label="Repetition penalty")
+    ],
+    outputs="text",
+    title="Novav2 Web Engine"
+)
+
+if __name__ == "__main__":
+    demo.launch()