Update app.py

app.py

@@ -1,6 +1,5 @@
 import chainlit as cl
 from langchain.embeddings.openai import OpenAIEmbeddings
-from langchain.document_loaders.csv_loader import CSVLoader
 from langchain.embeddings import CacheBackedEmbeddings
 from langchain.text_splitter import RecursiveCharacterTextSplitter
 from langchain.vectorstores import FAISS
@@ -12,23 +11,11 @@ from langchain.prompts.chat import (
     SystemMessagePromptTemplate,
     HumanMessagePromptTemplate,
 )
-import chainlit as cl
 
 text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=100)
 
 system_template = """
-Use the following pieces of context to answer the user's question.
-Please respond as if you were Ken from the movie Barbie. Ken is a well-meaning but naive character who loves to Beach. He talks like a typical Californian Beach Bro, but he doesn't use the word "Dude" so much.
-If you don't know the answer, just say that you don't know, don't try to make up an answer.
-You can make inferences based on the context as long as it still faithfully represents the feedback.
-
-Example of your response should be:
-
-```
-The answer is foo
-```
-
-Begin!
+Use the following pieces of context to answer the user's question. If the question cannot be answered with the supplied context, simply answer "I cannot determine this based on the provided context."  
 ----------------
 {context}"""
 
@@ -41,7 +28,7 @@ chain_type_kwargs = {"prompt": prompt}
 
 @cl.author_rename
 def rename(orig_author: str):
-    rename_dict = {"RetrievalQA": "Consulting The Kens"}
+    rename_dict = {"RetrievalQA": "Consulting PageTurn"}
     return rename_dict.get(orig_author, orig_author)
 
 @cl.on_chat_start
@@ -49,16 +36,21 @@ async def init():
     msg = cl.Message(content=f"Building Index...")
     await msg.send()
 
-    # build FAISS index from csv
-    loader = CSVLoader(file_path="./data/barbie.csv", source_column="Review_Url")
-    data = loader.load()
-    documents = text_splitter.transform_documents(data)
+    # Read text from a .txt file
+    with open('./data/aerodynamic_drag.txt', 'r') as f:
+        aerodynamic_drag_data = f.read()
+
+    # Split the text into smaller chunks
+    documents = text_splitter.transform_documents([aerodynamic_drag_data])
+
+    # Create a local file store for caching
     store = LocalFileStore("./cache/")
     core_embeddings_model = OpenAIEmbeddings()
     embedder = CacheBackedEmbeddings.from_bytes_store(
         core_embeddings_model, store, namespace=core_embeddings_model.model
     )
-    # make async docsearch
+
+    # Make async docsearch
     docsearch = await cl.make_async(FAISS.from_documents)(documents, embedder)
 
     chain = RetrievalQA.from_chain_type(
@@ -66,7 +58,7 @@ async def init():
         chain_type="stuff",
         return_source_documents=True,
         retriever=docsearch.as_retriever(),
-        chain_type_kwargs = {"prompt": prompt}
+        chain_type_kwargs={"prompt": prompt}
     )
 
     msg.content = f"Index built!"
@@ -74,7 +66,6 @@ async def init():
 
     cl.user_session.set("chain", chain)
 
-
 @cl.on_message
 async def main(message):
     chain = cl.user_session.get("chain")