Update app.py

app.py

@@ -14,8 +14,35 @@ df['embeding_context'] = df['embeding_context'].astype(str).fillna('')
 
 # Filter out any rows where 'embeding_context' might be empty or invalid
 df = df[df['embeding_context'] != '']
-    
-index = faiss.read_index('vector_store.index')
+
+# Encode the 'embeding_context' column
+embedding_contexts = df['embeding_context'].tolist()
+embeddings_csv = model.encode(embedding_contexts, batch_size=12, max_length=1024)['dense_vecs']
+
+# Convert embeddings to numpy array
+embeddings_np = np.array(embeddings_csv).astype('float32')
+
+# FAISS index file path
+index_file_path = 'vector_store_bge_m3.index'
+
+# Check if FAISS index file already exists
+if os.path.exists(index_file_path):
+    # Load the existing FAISS index from file
+    index = faiss.read_index(index_file_path)
+    print("FAISS index loaded from file.")
+else:
+    # Initialize FAISS index (for L2 similarity)
+    dim = embeddings_np.shape[1]
+    index = faiss.IndexFlatL2(dim)
+
+    # Add embeddings to the FAISS index
+    index.add(embeddings_np)
+
+    # Save the FAISS index to a file for future use
+    faiss.write_index(index, index_file_path)
+    print("FAISS index created and saved to file.")
+
+index = faiss.read_index(index_file_path)
 
 
 # Function to perform search and return all columns
@@ -23,7 +50,7 @@ def search_query(query_text):
     num_records = 50
 
     # Encode the input query text
-    embeddings_query = model.encode([query_text], batch_size=12, max_length=2048)['dense_vecs']
+    embeddings_query = model.encode([query_text], batch_size=12, max_length=1024)['dense_vecs']
     embeddings_query_np = np.array(embeddings_query).astype('float32')
 
     # Search in FAISS index for nearest neighbors
@@ -41,16 +68,16 @@ def gradio_interface(query_text):
 
 with gr.Blocks() as app:
     gr.Markdown("<h1>White Stride Red Search (BEG-M3)</h1>")
-    
+
     # Input text box for the search query
     search_input = gr.Textbox(label="Search Query", placeholder="Enter search text", interactive=True)
-    
+
     # Search button below the text box
     search_button = gr.Button("Search")
-    
+
     # Output table for displaying results
     search_output = gr.DataFrame(label="Search Results")
-    
+
     # Link button click to action
     search_button.click(fn=gradio_interface, inputs=search_input, outputs=search_output)