Update app.py

app.py

@@ -66,6 +66,7 @@ def get_text_embedding(text):
         text_features = clip_model.encode_text(text_tokens)
         text_features /= text_features.norm(dim=-1, keepdim=True)
     return text_features.cpu().numpy()
+    
 def get_average_embedding(main_image_url, additional_image_urls):
     embeddings = []
     
@@ -82,7 +83,7 @@ def get_average_embedding(main_image_url, additional_image_urls):
     
     if embeddings:
         avg_embedding = np.mean(embeddings, axis=0)
-        return avg_embedding.tolist()  # numpy 배열을 파이썬 리스트로 변환
+        return avg_embedding.tolist() if isinstance(avg_embedding, np.ndarray) else avg_embedding
     else:
         return None
 
@@ -105,7 +106,7 @@ def update_collection_embeddings():
             try:
                 avg_embedding = get_average_embedding(main_image_url, additional_image_urls)
                 if avg_embedding is not None:
-                    batch_embeddings.append(avg_embedding)  # 이미 리스트 형태임
+                    batch_embeddings.append(avg_embedding)
                     valid_ids.append(id)
                 else:
                     st.warning(f"Failed to generate embedding for item {id}")
@@ -123,9 +124,12 @@ def update_collection_embeddings():
                 st.error(f"Error updating embeddings: {str(e)}")
                 st.error(f"First embedding type: {type(batch_embeddings[0])}")
                 st.error(f"First embedding length: {len(batch_embeddings[0])}")
+                st.error(f"First embedding: {batch_embeddings[0][:10]}...")  # 처음 10개 요소만 출력
         
         # 진행 상황 표시
         st.progress((i + batch_size) / len(all_ids))
+
+        
 def find_similar_images(query_embedding, collection, top_k=5):
     results = collection.query(
         query_embeddings=[query_embedding.squeeze().tolist()],