caching the data loading

app.py

@@ -6,19 +6,24 @@ from haystack.nodes import FARMReader, TfidfRetriever
 import validators
 import json
 
-doc_dir = './article_txt_got'
+#Haystack Components
 document_store = InMemoryDocumentStore()
-docs = convert_files_to_docs(dir_path=doc_dir, clean_func=clean_wiki_text, split_paragraphs=True)
-
-document_store.write_documents(docs)
 retriever = TfidfRetriever(document_store=document_store)
 reader = FARMReader(model_name_or_path="deepset/roberta-base-squad2", use_gpu=True)
-
 pipeline = ExtractiveQAPipeline(reader, retriever)
 
+@st.cache
+def load_and_write_data():
+    doc_dir = './article_txt_got'
+    docs = convert_files_to_docs(dir_path=doc_dir, clean_func=clean_wiki_text, split_paragraphs=True)
+
+    document_store.write_documents(docs)
+
 #Streamlit App
 
-st.set_page_config(page_title='Game of Thrones QA with Haystack')
+st.title('Game of Thrones QA with Haystack')
+
+load_and_write_data()
 
 def ask_question(question):
     prediction = pipeline.run(query=question, params={"Retriever": {"top_k": 10}, "Reader": {"top_k": 5}})