File size: 3,958 Bytes
623e43b 29fca42 623e43b 29fca42 |
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 |
import streamlit as st
import pandas as pd
import tempfile
import os
from langchain.document_loaders import DataFrameLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.embeddings import HuggingFaceEmbeddings
from langchain.vectorstores import FAISS
from langchain.chains import RetrievalQA
from langchain import HuggingFacePipeline
from transformers import pipeline, AutoTokenizer, AutoModelForSeq2SeqLM
def preprocess_excel(file_path: str) -> pd.DataFrame:
df_raw = pd.read_excel(file_path, sheet_name='Data Base', header=None)
df = df_raw.iloc[4:].copy()
df.columns = df.iloc[0]
df = df[1:]
df.dropna(how='all', inplace=True)
df.dropna(axis=1, how='all', inplace=True)
df.reset_index(drop=True, inplace=True)
return df
def build_vectorstore_from_dataframe(df: pd.DataFrame):
df.fillna("", inplace=True)
df['combined_text'] = df.apply(lambda row: ' | '.join([str(cell) for cell in row]), axis=1)
docs_loader = DataFrameLoader(df[['combined_text']], page_content_column='combined_text')
documents = docs_loader.load()
splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=150)
split_docs = splitter.split_documents(documents)
embeddings = HuggingFaceEmbeddings(
model_name="sentence-transformers/all-MiniLM-l6-v2",
model_kwargs={"device": "cpu"},
encode_kwargs={"normalize_embeddings": False}
)
vectorstore = FAISS.from_documents(split_docs, embeddings)
return vectorstore
def create_qa_pipeline(vectorstore):
model_id = "google/flan-t5-base"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForSeq2SeqLM.from_pretrained(model_id)
gen_pipeline = pipeline("text2text-generation", model=model, tokenizer=tokenizer, max_length=512)
llm = HuggingFacePipeline(pipeline=gen_pipeline)
retriever = vectorstore.as_retriever()
qa = RetrievalQA.from_chain_type(llm=llm, retriever=retriever, chain_type="stuff", return_source_documents=False)
return qa
st.set_page_config(page_title="Excel-Aware RAG Chatbot", layout="wide")
st.title("📊 Excel-Aware RAG Chatbot (Professional QA)")
with st.sidebar:
uploaded_file = st.file_uploader("Upload your Excel file (.xlsx or .xlsm with 'Data Base' sheet)", type=["xlsx", "xlsm"])
if uploaded_file is not None:
with st.spinner("Processing and indexing your Excel sheet..."):
with tempfile.NamedTemporaryFile(delete=False, suffix=".xlsm") as tmp_file:
tmp_file.write(uploaded_file.read())
tmp_path = tmp_file.name
try:
cleaned_df = preprocess_excel(tmp_path)
vectorstore = build_vectorstore_from_dataframe(cleaned_df)
qa = create_qa_pipeline(vectorstore)
st.success("✅ File processed and chatbot ready! Ask your questions below.")
if "chat_history" not in st.session_state:
st.session_state.chat_history = []
with st.chat_message("assistant"):
st.markdown("How can I help you with the inspection data?")
user_prompt = st.chat_input("Ask a question like 'How many backlog items are marked Yes?' or 'List overdue inspections'.")
if user_prompt:
st.chat_message("user").markdown(user_prompt)
with st.chat_message("assistant"):
with st.spinner("Thinking..."):
try:
answer = qa.run(user_prompt)
st.markdown(f"**Answer:** {answer}")
st.session_state.chat_history.append((user_prompt, answer))
except Exception as e:
st.error(f"Error: {e}")
except Exception as e:
st.error(f"Error processing file: {e}")
finally:
os.remove(tmp_path)
else:
st.info("Upload a file to get started.")
|