Create store_index.py

store_index.py

@@ -0,0 +1,28 @@
+from src.helper import load_pdf, text_split, download_hugging_face_embeddings
+from langchain.vectorstores import FAISS
+from dotenv import load_dotenv
+from langchain.schema import Document
+
+load_dotenv()
+
+# Load and process PDF data
+extracted_data = load_pdf("data/")
+text_chunks = text_split(extracted_data)
+
+# Download embeddings model
+embedding_model = download_hugging_face_embeddings()
+
+# Extract the page contents from the text chunks
+texts = [chunk.page_content for chunk in text_chunks]
+
+# Generate embeddings for the text chunks
+embeddings = embedding_model.embed_documents(texts)
+
+# Create Document objects with page content and embeddings
+documents = [Document(page_content=text, embedding=embedding) for text, embedding in zip(texts, embeddings)]
+
+# Initialize FAISS vector store with documents
+vector_store = FAISS.from_documents(documents, embedding_model)
+
+# Save the vector store to disk for later use
+vector_store.save_local("vector_store")