Update app.py

app.py

@@ -12,6 +12,7 @@ from huggingface_hub import login
 import arxiv
 import numpy as np
 import torch  # Add torch to explicitly set the device
+import gradio as gr
 
 # Access the Hugging Face token from the environment variable
 HF_TOKEN = os.getenv("HF_Token")
@@ -63,7 +64,6 @@ def download_youtube_video(video_url, output_dir, title=None):
     try:
         with YoutubeDL(ydl_opts) as ydl:
             info = ydl.extract_info(video_url, download=True)
-            # Extract the final downloaded file path
             downloaded_file = ydl.prepare_filename(info)
         return downloaded_file
     except Exception as e:
@@ -81,22 +81,18 @@ def fetch_and_download_youtube_video(query, output_dir="./videos"):
 
     try:
         with YoutubeDL(ydl_opts) as ydl:
-            # Perform a search for the query on YouTube
             search_results = ydl.extract_info(f"ytsearch:{query}", download=False)
-            
             if 'entries' not in search_results or len(search_results['entries']) == 0:
                 print(f"No YouTube results found for query: '{query}'")
                 return []
 
-            video_info = search_results['entries'][0]  # Take the first result
+            video_info = search_results['entries'][0]
             video_title = video_info.get("title", "unknown_title")
             video_url = video_info.get("webpage_url", None)
-            
             if not video_url:
                 print("No URL found for the video.")
                 return []
 
-            # Download the video
             local_path = download_youtube_video(video_url, output_dir, title=video_title)
             if not local_path:
                 return []
@@ -108,24 +104,21 @@ def fetch_and_download_youtube_video(query, output_dir="./videos"):
         print(f"Error fetching YouTube video for query '{query}': {e}")
         return []
 
-from arxiv import Client, Search, SortCriterion
-
 def fetch_from_arxiv(query="machine learning", max_results=2, output_dir="./papers"):
     """Fetch papers from arXiv and download their PDFs."""
     print(f"Fetching papers for query: {query}")
-    client = Client()
-    search = Search(
+    client = arxiv.Client()
+    search = arxiv.Search(
         query=query,
         max_results=max_results,
-        sort_by=SortCriterion.Relevance
+        sort_by=arxiv.SortCriterion.Relevance
     )
     metadata = []
     for i, result in enumerate(client.results(search)):
-        pdf_url = result.pdf_url  # Direct link to PDF
+        pdf_url = result.pdf_url
         filename = f"{query.replace(' ', '_')}_arxiv_{i}.pdf"
         local_path = os.path.join(output_dir, filename)
         try:
-            # Download the PDF
             response = requests.get(pdf_url)
             if response.status_code == 200:
                 with open(local_path, 'wb') as f:
@@ -182,8 +175,21 @@ def hybrid_rag_system_with_llama(query):
     final_response = generate_llama_response(query, context)
     return final_results, final_response
 
-# Example query
-query = "short easy machine learning"
-results, final_response = hybrid_rag_system_with_llama(query)
-print("\nFinal Response Generated by Llama 3:")
-print(final_response)
+# Define Gradio interface
+def gradio_interface(query):
+    """Gradio wrapper for hybrid RAG system."""
+    _, final_response = hybrid_rag_system_with_llama(query)
+    return final_response
+
+# Create Gradio app
+interface = gr.Interface(
+    fn=gradio_interface,
+    inputs=gr.Textbox(label="Enter your query", placeholder="e.g., short easy machine learning"),
+    outputs=gr.Textbox(label="Generated Response"),
+    title="Hybrid RAG System with LLaMA",
+    description="Enter a query to retrieve relevant resources and generate a response using LLaMA."
+)
+
+# Launch Gradio app
+if __name__ == "__main__":
+    interface.launch()