updated app.py

app.py

@@ -120,15 +120,15 @@ def create_book_embeddings(book_content):
     
     
     print("Creating book embeddings...")
-    # try:
-    #     book_embeddings = Chroma.load(persist_directory = '.',
-    #                            collection_name = 'book')
-    # except:
-    book_embeddings = Chroma.from_documents(documents = texts,
-                                        embedding = instructor_embeddings,
-                                        persist_directory = '.',
-                                        collection_name = 'book')
-    print("Book embeddings created.")
+    try:
+        book_embeddings = Chroma.load(persist_directory = '.',
+                               collection_name = 'book')
+    except:
+        book_embeddings = Chroma.from_documents(documents = texts,
+                                            embedding = instructor_embeddings,
+                                            persist_directory = '.',
+                                            collection_name = 'book')
+        print("Book embeddings created.")
         
     book_embeddings.add_documents(documents=texts, embedding=instructor_embeddings)
     book_embeddings.persist()
@@ -164,7 +164,7 @@ PROMPT = PromptTemplate(
 )
 
 
-def generate_answer_from_embeddings(query, book_embeddings):
+def generate_answer_from_embeddings(query):
     """
     Retrieve documents from the vector database and then pass them to the language model to generate an answer.
 
@@ -215,4 +215,4 @@ def get_book(book_name: str):
 
 @app.get("/answer")
 def get_answer(query: str):
-    return generate_answer_from_embeddings(query, book_embeddings)
+    return generate_answer_from_embeddings(query)