Spaces:
Runtime error
Runtime error
File size: 4,160 Bytes
2aea9eb 77d8ee7 2aea9eb dc22f3f 2aea9eb dc22f3f 2aea9eb 344a7e8 2aea9eb 344a7e8 2aea9eb 344a7e8 2aea9eb dc4d491 2aea9eb dc22f3f 2aea9eb dc22f3f 2aea9eb 0684c75 4965282 2aea9eb | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 | import os
import streamlit as st
import pandas as pd
from langchain_community.vectorstores import FAISS
from langchain_openai import OpenAIEmbeddings
from langchain_openai.chat_models import ChatOpenAI
from langchain.memory import ConversationBufferMemory
from langchain.chains import ConversationalRetrievalChain
from langchain.text_splitter import CharacterTextSplitter
from time import sleep
OPENAI_API_KEY = "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"
os.environ['OPENAI_API_KEY'] = OPENAI_API_KEY
# Initialize variables
vectorstore = None
conversation_chain = None
chat_history = []
# Function to process uploaded CSV file
def process_csv(csv_file):
try:
df = pd.read_csv(csv_file, encoding='latin-1') # Specify encoding as latin-1 or ISO-8859-1
text = df.to_string(index=False)
text_chunks = get_text_chunks(text)
vectorstore = get_vectorstore(text_chunks)
if vectorstore:
conversation_chain = get_conversation_chain(vectorstore)
return conversation_chain
else:
st.error("Failed to create vectorstore. Rate limit exceeded. Please try again later.")
return None
except Exception as e:
st.error(f"Error processing CSV file: {e}")
return None
# Function to split text into chunks
def get_text_chunks(text):
text_splitter = CharacterTextSplitter(
separator="\n",
chunk_size=2000, # Increased chunk size for larger datasets
chunk_overlap=40, # Increased chunk overlap for larger datasets
length_function=len
)
chunks = text_splitter.split_text(text)
return chunks
# Function to create vectorstore from text chunks
def get_vectorstore(text_chunks):
retries = 5 # Increased number of retries for larger datasets
for i in range(retries):
try:
embeddings = OpenAIEmbeddings()
vectorstore = FAISS.from_texts(texts=text_chunks, embedding=embeddings)
return vectorstore
except Exception as e:
st.warning(f"Retry {i+1}/{retries}: Waiting for 20 seconds due to rate limit exceeded.")
sleep(20)
return None
# Function to create conversation chain
def get_conversation_chain(vectorstore):
llm = ChatOpenAI()
memory = ConversationBufferMemory(
memory_key='chat_history', return_messages=True)
conversation_chain = ConversationalRetrievalChain.from_llm(
llm=llm,
retriever=vectorstore.as_retriever(), # This might still raise an error if vectorstore is None
memory=memory
)
return conversation_chain
# Streamlit app
def main():
global vectorstore, conversation_chain, chat_history
st.title('CSV Chatbot')
# Page to upload CSV file
st.subheader('Upload CSV File')
csv_file = st.file_uploader('Upload CSV', type=['csv'])
if csv_file:
conversation_chain = process_csv(csv_file)
if conversation_chain:
# Chat interface
st.subheader('Chat Interface')
user_question = st.text_input('Ask a question:')
if st.button('Ask'):
# Split the user question into smaller parts
questions = user_question.split('?')
for question in questions:
response = conversation_chain.invoke({'question': question.strip()})
if 'chat_history' in response:
chat_history = response['chat_history']
for message in chat_history:
if isinstance(message, dict) and 'role' in message and 'content' in message:
if message['role'] == 'user':
st.write(f"You: {message['content']}")
elif message['role'] == 'assistant':
st.write(f"Assistant: {message['content']}")
else:
st.error("Failed to get response. Please try again.")
else:
st.error("Failed to process CSV file. Please try again.")
if __name__ == '__main__':
main()
|