Update app.py
Browse files
app.py
CHANGED
|
@@ -64,11 +64,8 @@ def load_and_index_pdf(pdf_path):
|
|
| 64 |
logging.info(f"Loading PDF document from: {pdf_path}")
|
| 65 |
documents = SimpleDirectoryReader(input_files=[pdf_path]).load_data()
|
| 66 |
if documents:
|
| 67 |
-
logging.info("Creating semantic splitter")
|
| 68 |
-
node_parser = create_semantic_splitter_node_parser()
|
| 69 |
-
nodes = node_parser.get_nodes_from_documents(documents)
|
| 70 |
logging.info("Creating vector store index")
|
| 71 |
-
index = VectorStoreIndex(
|
| 72 |
return index
|
| 73 |
else:
|
| 74 |
logging.warning("No documents found in the PDF")
|
|
@@ -91,10 +88,8 @@ def create_rag_pipeline(index):
|
|
| 91 |
logging.info("Initializing RAG Pipeline components")
|
| 92 |
# setup retriever
|
| 93 |
|
| 94 |
-
retriever =
|
| 95 |
-
|
| 96 |
-
).as_retriever(similarity_top_k=5)
|
| 97 |
-
|
| 98 |
|
| 99 |
# setup query transformer
|
| 100 |
hyde_query_transform = HyDEQueryTransform(llm=Settings.llm)
|
|
|
|
| 64 |
logging.info(f"Loading PDF document from: {pdf_path}")
|
| 65 |
documents = SimpleDirectoryReader(input_files=[pdf_path]).load_data()
|
| 66 |
if documents:
|
|
|
|
|
|
|
|
|
|
| 67 |
logging.info("Creating vector store index")
|
| 68 |
+
index = VectorStoreIndex.from_documents(documents, node_parser=create_semantic_splitter_node_parser())
|
| 69 |
return index
|
| 70 |
else:
|
| 71 |
logging.warning("No documents found in the PDF")
|
|
|
|
| 88 |
logging.info("Initializing RAG Pipeline components")
|
| 89 |
# setup retriever
|
| 90 |
|
| 91 |
+
retriever = index.as_retriever(similarity_top_k=5)
|
| 92 |
+
|
|
|
|
|
|
|
| 93 |
|
| 94 |
# setup query transformer
|
| 95 |
hyde_query_transform = HyDEQueryTransform(llm=Settings.llm)
|