Update app.py

app.py

@@ -1,11 +1,24 @@
+import os
 import gradio as gr
 from huggingface_hub import InferenceClient
+import markdown
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
 client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 
+# Function to read and process Markdown files from the 'data' directory
+def load_markdown_files(data_folder='data'):
+    documents = []
+    for filename in os.listdir(data_folder):
+        if filename.endswith('.md'):
+            with open(os.path.join(data_folder, filename), 'r', encoding='utf-8') as file:
+                content = file.read()
+                # Convert Markdown to plain text if needed
+                html_content = markdown.markdown(content)
+                documents.append(html_content)  # Store HTML content or plain text
+    return documents
+
+# Load documents at startup
+documents = load_markdown_files()
 
 def respond(
     message,
@@ -17,6 +30,12 @@ def respond(
 ):
     messages = [{"role": "system", "content": system_message}]
 
+    # Retrieve relevant context from loaded documents based on the user message
+    relevant_contexts = retrieve_relevant_context(message, documents)
+
+    # Add retrieved contexts to the messages for better responses
+    messages.append({"role": "context", "content": " ".join(relevant_contexts)})
+
     for val in history:
         if val[0]:
             messages.append({"role": "user", "content": val[0]})
@@ -35,13 +54,19 @@ def respond(
         top_p=top_p,
     ):
         token = message.choices[0].delta.content
-
         response += token
         yield response
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
+def retrieve_relevant_context(query, documents):
+    # Simple keyword matching to find relevant documents
+    relevant_contexts = []
+    
+    for doc in documents:
+        if query.lower() in doc.lower():  # Basic keyword search (case insensitive)
+            relevant_contexts.append(doc)
+
+    return relevant_contexts[:3]  # Return top 3 relevant contexts
+
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[
@@ -58,6 +83,5 @@ demo = gr.ChatInterface(
     ],
 )
 
-
 if __name__ == "__main__":
     demo.launch()