Spaces:
Running
Running
File size: 5,798 Bytes
40e0a99 0e95be6 40e0a99 0e95be6 40e0a99 0e95be6 40e0a99 0e95be6 40e0a99 0e95be6 40e0a99 0e95be6 40e0a99 0e95be6 40e0a99 0e95be6 40e0a99 0e95be6 40e0a99 0e95be6 40e0a99 0e95be6 40e0a99 0e95be6 40e0a99 0e95be6 40e0a99 0e95be6 40e0a99 0e95be6 40e0a99 0e95be6 40e0a99 0e95be6 40e0a99 0e95be6 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 |
import streamlit as st
import pandas as pd
import os
import traceback
from dotenv import load_dotenv
from llama_index.readers.file.paged_csv.base import PagedCSVReader
from llama_index.core import Settings, VectorStoreIndex
from llama_index.llms.openai import OpenAI
from llama_index.embeddings.openai import OpenAIEmbedding
from llama_index.vector_stores.faiss import FaissVectorStore
from langchain_community.vectorstores import FAISS as LangChainFAISS
from langchain_community.docstore.in_memory import InMemoryDocstore
from langchain.chains import create_retrieval_chain
from langchain.chains.combine_documents import create_stuff_documents_chain
from langchain_core.prompts import ChatPromptTemplate
from langchain_openai import OpenAIEmbeddings, ChatOpenAI
from langchain_core.documents import Document
import faiss
import tempfile
# Load environment variables
os.environ["OPENAI_API_KEY"] = os.getenv("OPENAI_API_KEY")
# ✅ Check OpenAI API Key
if not os.getenv("OPENAI_API_KEY"):
st.error("⚠️ OpenAI API Key is missing! Please check your .env file or environment variables.")
# ✅ Ensure OpenAI Embeddings match FAISS dimensions
embedding_function = OpenAIEmbeddings()
test_vector = embedding_function.embed_query("test")
faiss_dimension = len(test_vector)
# ✅ Update global settings for LlamaIndex
Settings.llm = OpenAI(model="gpt-4o")
Settings.embed_model = OpenAIEmbedding(model="text-embedding-3-small", dimensions=faiss_dimension)
# Streamlit app
st.title("Chat with CSV Files - LangChain vs LlamaIndex")
# File uploader
uploaded_file = st.file_uploader("Upload a CSV file", type=["csv"])
if uploaded_file:
try:
# Read and preview CSV data using pandas
data = pd.read_csv(uploaded_file)
st.write("Preview of uploaded data:")
st.dataframe(data)
# Save the uploaded file to a temporary location
with tempfile.NamedTemporaryFile(delete=False, suffix=".csv", mode="w", encoding="utf-8") as temp_file:
temp_file_path = temp_file.name
data.to_csv(temp_file.name, index=False, encoding="utf-8")
temp_file.flush()
# Tabs for LangChain and LlamaIndex
tab1, tab2 = st.tabs(["Chat w CSV using LangChain", "Chat w CSV using LlamaIndex"])
# ✅ LangChain Processing
with tab1:
st.subheader("LangChain Query")
try:
# ✅ Store each row as a single document
st.write("Processing CSV with a custom loader...")
documents = []
for _, row in data.iterrows():
content = " | ".join([f"{col}: {row[col]}" for col in data.columns]) # ✅ Store entire row as a document
doc = Document(page_content=content)
documents.append(doc)
# ✅ Create FAISS VectorStore
st.write(f"✅ Initializing FAISS with dimension: {faiss_dimension}")
langchain_index = faiss.IndexFlatL2(faiss_dimension)
docstore = InMemoryDocstore()
index_to_docstore_id = {}
langchain_vector_store = LangChainFAISS(
embedding_function=embedding_function,
index=langchain_index,
docstore=docstore,
index_to_docstore_id=index_to_docstore_id,
)
# ✅ Ensure documents are added correctly
try:
langchain_vector_store.add_documents(documents)
st.write("✅ Documents successfully added to FAISS VectorStore.")
except Exception as e:
st.error(f"Error adding documents to FAISS: {e}")
st.text(traceback.format_exc())
# ✅ Limit number of retrieved documents
retriever = langchain_vector_store.as_retriever(search_kwargs={"k": 15}) # Fetch 15 docs instead of 5
# ✅ Query Processing
query = st.text_input("Ask a question about your data (LangChain):")
if query:
try:
retrieved_docs = retriever.get_relevant_documents(query)
retrieved_context = "\n\n".join([doc.page_content for doc in retrieved_docs])
retrieved_context = retrieved_context[:3000]
# ✅ Show retrieved context for debugging
st.write("🔍 **Retrieved Context Preview:**")
st.text(retrieved_context)
system_prompt = (
"You are an assistant for question-answering tasks. "
"Use the following pieces of retrieved context to answer "
"the question. Keep the answer concise.\n\n"
f"{retrieved_context}"
)
# Simulate LangChain RAG Chain (update actual logic if necessary)
st.write("🚀 Query processed successfully.")
st.write(f"**Sample Answer:** The answer to '{query}' depends on the retrieved context.")
except Exception as e:
error_message = traceback.format_exc()
st.error(f"Error processing query: {e}")
st.text(error_message)
except Exception as e:
error_message = traceback.format_exc()
st.error(f"Error processing with LangChain: {e}")
st.text(error_message)
except Exception as e:
error_message = traceback.format_exc()
st.error(f"Error reading uploaded file: {e}")
st.text(error_message)
|