Upload folder using huggingface_hub

.gitignore

@@ -1,3 +1,4 @@
 .venv
 .env
-*pycache*
+*pycache*
+voc_bot

front_end.py

@@ -3,22 +3,45 @@ from rag import mongo_rag_tool
 from gradio.themes.base import Base
 
 # Create an instance of GradIO
-
-
 with gr.Blocks(theme=Base(), title="Market Research and VOC bot") as demo:
+    # A styled header for the app
+    gr.Markdown(
+        """
+        <div style='text-align: center; font-size: 24px; font-weight: bold; margin-bottom: 20px;'>
+        Chat with your data
+        </div>
+        """
+    )
+    
+    # Input fields for the collection and the question, with descriptive text
     gr.Markdown(
         """
-        # VOC App using mined data
-        """)
-    textbox = gr.Textbox(label="Enter your Question:")
+        <div style='text-align: left; font-size: 18px; margin-bottom: 10px;'>
+        Enter the collection and your query to get relevant answers:
+        </div>
+        """
+    )
+    collection_textbox = gr.Textbox(label="Enter your Collection:", placeholder="e.g., market_data", lines=1)
+    query_textbox = gr.Textbox(label="Enter your Question:", placeholder="Type your question here...", lines=1)
+
+    # Submit button with some spacing and central alignment
     with gr.Row():
-        button = gr.Button("Submit", variant="primary")
-    with gr.Column():
-        output1 = gr.Textbox(lines=1, max_lines=10, label="Answer:")
-        output2 = gr.Textbox(lines=1, max_lines=10, label="Sources:")
+        button = gr.Button("Submit", variant="primary", size="lg")
 
-#    Call query_data function upon clicking the Submit button
+    # Output section for displaying the answer and sources one below the other
+    gr.Markdown(
+        """
+        <div style='text-align: left; font-size: 18px; margin-bottom: 10px;'>
+        Output:
+        </div>
+        """
+    )
+    # Using a Column to place the answer and source outputs one after the other
+    with gr.Column():
+        answer_output = gr.Textbox(lines=5, label="Answer:", max_lines=50)
+        source_output = gr.Textbox(lines=5, label="Sources:", max_lines=50)
 
-    button.click(mongo_rag_tool, textbox, outputs=[output1, output2])
+    # Connect the button to the function
+    button.click(mongo_rag_tool, inputs=[query_textbox, collection_textbox], outputs=[answer_output, source_output])
 
-demo.launch()
+demo.launch()

rag.py

@@ -13,19 +13,18 @@ load_dotenv()
 INDEX_NAME = "vector_index"
 DATABASE_NAME = "scraped_data_db"
 
-def mongo_rag_tool(query: str) -> str:
+def mongo_rag_tool(query: str, collection_name: str) -> str:
     """
     This function is used to retrieve documents from a MongoDB database and then use the RAG model to answer the query.
     The documents that are most semantically close to the query are returned.
     args:
         query: str: The query that you want to use to retrieve documents
         collection_name: str: The name of the collection in the MongoDB database
-        output_filename: str: The name of the output file where the results will be saved
     returns:
         str: The answer to the query
     """
     try:
-        collection_name = os.getenv("MONGODB_COLLECTION_NAME")
+        #collection_name = os.getenv("MONGODB_COLLECTION_NAME")
         # Connect to the MongoDB database
         openai_api_key = os.getenv("OPENAI_API_KEY")
         embeddings = OpenAIEmbeddings(openai_api_key=openai_api_key, disallowed_special=(), model="text-embedding-3-small")