Update app.py

app.py

@@ -681,22 +681,74 @@ def generate_html(local_files):
     html += "</ul>"
     return html
 
+
+
+
+
+
+
+from gradio_client import Client
+client = Client("awacke1/Arxiv-Paper-Search-And-QA-RAG-Pattern")
+result = client.predict(
+		message="Hello!!",
+		llm_results_use=5,
+		database_choice="Semantic Search",
+		llm_model_picked="mistralai/Mistral-7B-Instruct-v0.2",
+		api_name="/update_with_rag_md"
+)
+print(result)
+Accepts 4 parameters:
+message str Required
+The input value that is provided in the "Search" Textbox component.
+llm_results_use float Default: 5
+The input value that is provided in the "Top n results as context" Slider component.
+database_choice Literal['Semantic Search', 'Arxiv Search - Latest - (EXPERIMENTAL)'] Default: "Semantic Search"
+The input value that is provided in the "Search Source" Dropdown component.
+llm_model_picked Literal['mistralai/Mixtral-8x7B-Instruct-v0.1', 'mistralai/Mistral-7B-Instruct-v0.2', 'google/gemma-7b-it', 'None'] Default: "mistralai/Mistral-7B-Instruct-v0.2"
+The input value that is provided in the "LLM Model" Dropdown component.
+Returns tuple of 2 elements
+[0] str
+The output value that appears in the "value_14" Markdown component.
+[1] str
+The output value that appears in the "value_13" Textbox component.
+
+
+
+
+
+
+
+
+
+
 #@st.cache_resource
 def search_arxiv(query):
     start_time = time.strftime("%Y-%m-%d %H:%M:%S")
     client = Client("awacke1/Arxiv-Paper-Search-And-QA-RAG-Pattern")
     response1 = client.predict(
-        query,
-        20,
-        "Semantic Search - up to 10 Mar 2024",
-        "mistralai/Mixtral-8x7B-Instruct-v0.1",
-        api_name="/update_with_rag_md"
+		message="Hello!!",
+		llm_results_use=5,
+		database_choice="Semantic Search",
+		llm_model_picked="mistralai/Mistral-7B-Instruct-v0.2",
+		api_name="/update_with_rag_md"
     )
+    
+    #response1 = client.predict(
+    #    query,
+    #    20,
+    #    "Semantic Search - up to 10 Mar 2024",
+    #    "mistralai/Mixtral-8x7B-Instruct-v0.1",
+    #    api_name="/update_with_rag_md"
+    #)
     Question = '### 🔎 ' + query + '\r\n'  # Format for markdown display with links
     References =  response1[0]  
-    ReferenceLinks = extract_urls(References)
+    References2 =  response1[1]  
 
+    st.code(References, language="markdown")
+    st.code(References2, language="markdown")
     
+    ReferenceLinks = extract_urls(References)
+
     filename = generate_filename(query, "md")
     create_file(filename, query, References + ReferenceLinks, should_save)
     st.session_state.messages.append({"role": "assistant", "content": References + ReferenceLinks})