Update app.py

app.py

@@ -5,6 +5,7 @@ from PyPDF2 import PdfReader
 import gradio as gr
 from datasets import Dataset, load_from_disk
 from sentence_transformers import SentenceTransformer
+import numpy as np
 
 # Extract text from PDF
 def extract_text_from_pdf(pdf_path):
@@ -45,7 +46,19 @@ os.makedirs(index_path, exist_ok=True)
 # Save the dataset to disk and create an index
 dataset.save_to_disk(dataset_path)
 dataset = load_from_disk(dataset_path)
-dataset.add_faiss_index(column="embeddings").save(index_path)
+
+# Add FAISS index while addressing numpy object deprecation
+def add_faiss_index(dataset, column):
+    import faiss  # Make sure faiss is installed
+    embeddings = np.array(dataset[column])
+    dim = embeddings.shape[1]
+    index = faiss.IndexFlatL2(dim)
+    index.add(embeddings)
+    dataset.add_faiss_index(column=column)
+    return dataset
+
+dataset = add_faiss_index(dataset, column="embeddings")
+dataset.save(index_path)
 
 # Custom retriever
 def retrieve(query):