Update app.py

app.py

@@ -3,16 +3,19 @@ from sentence_transformers import SentenceTransformer
 import numpy as np
 from transformers import pipeline
 import logging
+import sys
 
-# Set up logging
-logging.basicConfig(level=logging.ERROR)
+# Set up logging and debug print to console
+logging.basicConfig(level=logging.ERROR, stream=sys.stdout)
 logger = logging.getLogger(__name__)
 
-# Use a lighter summarization model to reduce resource use
+# Use a lighter summarization model
 try:
-    summarizer = pipeline("summarization", model="sshleifer/distilbart-cnn-6-6")  # Lighter than bart-large-cnn
+    summarizer = pipeline("summarization", model="sshleifer/distilbart-cnn-6-6")
+    print("Summarizer loaded successfully")
 except Exception as e:
     logger.error(f"Summarizer loading failed: {e}")
+    print(f"Summarizer error: {e}")
     summarizer = None
 
 # Sample dataset
@@ -37,24 +40,31 @@ papers = [
 # Load sentence transformer
 try:
     model = SentenceTransformer('all-MiniLM-L6-v2')
+    print("SentenceTransformer loaded successfully")
 except Exception as e:
     logger.error(f"SentenceTransformer loading failed: {e}")
+    print(f"SentenceTransformer error: {e}")
     model = None
 
 # Pre-compute embeddings
 paper_embeddings = model.encode([paper["content"] for paper in papers], convert_to_tensor=True) if model else None
+if paper_embeddings is not None:
+    print("Embeddings computed successfully")
+else:
+    print("Embeddings computation failed")
 
 def semantic_search(query):
-    # Check if any component is missing
     if model is None or summarizer is None or paper_embeddings is None:
-        return {"title": "Error", "abstract": "Error", "summary": "Model or embeddings failed to load. Check logs.", "link": ""}
+        return {"title": "Error", "abstract": "Error", "summary": "Model or embeddings failed to load. Check console.", "link": ""}
     
     try:
-        query_embedding = model.encode([query], convert_to_tensor=True)  # Ensure list input
+        print(f"Processing query: {query}")
+        query_embedding = model.encode([query], convert_to_tensor=True)
         similarities = np.dot(paper_embeddings, query_embedding.T).cpu().numpy()
         top_idx = np.argmax(similarities)
         top_paper = papers[top_idx]
         summary = summarizer(top_paper["content"], max_length=30, min_length=10, do_sample=False)[0]["summary_text"]
+        print(f"Found paper: {top_paper['title']}")
         return {
             "title": top_paper["title"],
             "abstract": top_paper["abstract"],
@@ -63,6 +73,7 @@ def semantic_search(query):
         }
     except Exception as e:
         logger.error(f"Search failed: {e}")
+        print(f"Search error: {e}")
         return {"title": "Error", "abstract": "Error", "summary": str(e), "link": ""}
 
 def search_interface(query):