update app.py

app.py

@@ -26,17 +26,17 @@ def get_pdf_text(pdf_docs):
         temp_file.write(pdf_docs.getvalue())
         temp_file.seek(0)
         pdf_loader = PyPDFLoader(temp_file.name)
-        print('pdf_loader = ', pdf_loader)
+        # print('pdf_loader = ', pdf_loader)
         pdf_doc = pdf_loader.load()
-        print('pdf_doc = ',pdf_doc)
+        # print('pdf_doc = ',pdf_doc)
         return pdf_doc
 
 
-def get_text_chunks(text):
-    print('text = ',text)
+def get_text_chunks(documents):
+    
     text_splitter = RecursiveCharacterTextSplitter(
-        chunk_size = 256,
-        chunk_overlap = 50,
+        chunk_size = 1000,
+        chunk_overlap = 200,
         length_function= len
     )
     # text_splitter = CharacterTextSplitter(
@@ -45,9 +45,9 @@ def get_text_chunks(text):
     #     chunk_overlap=200,
     #     length_function=len
     # )
-    chunks = text_splitter.split_text(text)
-    print('chunks = ', chunks)
-    return chunks
+    documents = text_splitter.split_documents(documents)
+    print('documents = ', documents)
+    return documents
 
 
 def get_vectorstore(text_chunks):
@@ -58,7 +58,7 @@ def get_vectorstore(text_chunks):
     # embeddings = OpenAIEmbeddings()sentence-transformers/all-MiniLM-L6-v2
     # embeddings = HuggingFaceInstructEmbeddings(model_name="hkunlp/instructor-xl",
     #                                           model_kwargs={'device':'cpu'})
-    vectorstore = FAISS.from_texts(texts=text_chunks, embedding=embeddings)
+    vectorstore = FAISS.from_documents(texts=text_chunks, embedding=embeddings)
     # vectorstore = Chroma.from_texts(texts=text_chunks, embedding=embeddings)
 
     return vectorstore
@@ -186,7 +186,7 @@ def main():
 
 
                 # get the text chunks
-                text_chunks = get_text_chunks(raw_text)
+                text_chunks = get_text_chunks(doc_list)
 
                 # create vector store
                 vectorstore = get_vectorstore(text_chunks)