sunbal7's picture
Update app.py
66e139c verified
raw
history blame
5.57 kB
import streamlit as st
st.set_page_config(page_title="RAG Book Analyzer", layout="wide") # Must be the first Streamlit command
import torch
import numpy as np
import faiss
from transformers import AutoModelForCausalLM, AutoTokenizer
from sentence_transformers import SentenceTransformer
import fitz # PyMuPDF for PDF extraction
import docx2txt # For DOCX extraction
from langchain_text_splitters import RecursiveCharacterTextSplitter
# ------------------------
# Configuration
# ------------------------
MODEL_NAME = "microsoft/phi-2" # Open-source model with good performance
EMBED_MODEL = "sentence-transformers/all-MiniLM-L6-v2" # Smaller embedding model
CHUNK_SIZE = 512
CHUNK_OVERLAP = 64
DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
# ------------------------
# Model Loading with Caching
# ------------------------
@st.cache_resource
def load_models():
try:
tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
MODEL_NAME,
device_map="auto" if DEVICE == "cuda" else None,
torch_dtype=torch.float16 if DEVICE == "cuda" else torch.float32,
trust_remote_code=True
)
embedder = SentenceTransformer(EMBED_MODEL, device=DEVICE)
return tokenizer, model, embedder
except Exception as e:
st.error(f"Model loading failed: {str(e)}")
st.stop()
tokenizer, model, embedder = load_models()
# ------------------------
# Text Processing Functions
# ------------------------
def split_text(text):
splitter = RecursiveCharacterTextSplitter(
chunk_size=CHUNK_SIZE,
chunk_overlap=CHUNK_OVERLAP,
length_function=len
)
return splitter.split_text(text)
def extract_text(file):
file_type = file.type
if file_type == "application/pdf":
try:
doc = fitz.open(stream=file.read(), filetype="pdf")
return "\n".join([page.get_text() for page in doc])
except Exception as e:
st.error("Error processing PDF: " + str(e))
return ""
elif file_type == "text/plain":
return file.read().decode("utf-8")
elif file_type == "application/vnd.openxmlformats-officedocument.wordprocessingml.document":
try:
return docx2txt.process(file)
except Exception as e:
st.error("Error processing DOCX: " + str(e))
return ""
else:
st.error("Unsupported file type: " + file_type)
return ""
def build_index(chunks):
embeddings = embedder.encode(chunks, show_progress_bar=False)
dimension = embeddings.shape[1]
index = faiss.IndexFlatL2(dimension)
index.add(embeddings)
return index
# ------------------------
# Summarization and Q&A Functions
# ------------------------
def generate_summary(text):
# Create prompt for Phi-2 model
prompt = f"Instruct: Summarize this book in a concise paragraph\nInput: {text[:3000]}\nOutput:"
inputs = tokenizer(prompt, return_tensors="pt").to(DEVICE)
outputs = model.generate(
**inputs,
max_new_tokens=300,
temperature=0.7,
top_p=0.9,
do_sample=True
)
summary = tokenizer.decode(outputs[0], skip_special_tokens=True)
return summary.split("Output:")[-1].strip()
def generate_answer(query, context):
# Create prompt for Phi-2 model
prompt = f"Instruct: Answer this question based on the context. If unsure, say 'I don't know'.\nQuestion: {query}\nContext: {context}\nOutput:"
inputs = tokenizer(prompt, return_tensors="pt").to(DEVICE)
outputs = model.generate(
**inputs,
max_new_tokens=300,
temperature=0.5,
top_p=0.9,
repetition_penalty=1.2,
do_sample=True
)
answer = tokenizer.decode(outputs[0], skip_special_tokens=True)
return answer.split("Output:")[-1].strip()
# ------------------------
# Streamlit UI
# ------------------------
st.title("πŸ“š RAG-Based Book Analyzer")
st.write("Upload a book (PDF, TXT, DOCX) to get a summary and ask questions about its content.")
uploaded_file = st.file_uploader("Upload File", type=["pdf", "txt", "docx"])
if uploaded_file:
text = extract_text(uploaded_file)
if text:
st.success("βœ… File successfully processed!")
with st.spinner("Generating summary..."):
summary = generate_summary(text)
st.markdown("### Book Summary")
st.info(summary)
# Process text into chunks and build FAISS index
chunks = split_text(text)
index = build_index(chunks)
st.session_state.chunks = chunks
st.session_state.index = index
st.markdown("### ❓ Ask a Question about the Book")
query = st.text_input("Enter your question:")
if query:
with st.spinner("Searching for answers..."):
# Retrieve top 3 relevant chunks as context
query_embedding = embedder.encode([query])
distances, indices = st.session_state.index.search(query_embedding, k=3)
retrieved_chunks = [st.session_state.chunks[i] for i in indices[0] if i < len(st.session_state.chunks)]
context = "\n\n".join(retrieved_chunks)
answer = generate_answer(query, context)
st.markdown("### πŸ’¬ Answer")
st.success(answer)
with st.expander("See context used"):
st.write(context)