File size: 2,142 Bytes
1506f8e d28742c 743388b 1506f8e d28742c 1506f8e 743388b d28742c 743388b d28742c 743388b d28742c 743388b 1506f8e 743388b 184a854 d28742c 743388b d28742c 743388b d28742c 1506f8e 743388b |
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 |
import streamlit as st
import PyPDF2
import os
from sentence_transformers import SentenceTransformer
import faiss
import numpy as np
from transformers import pipeline
st.set_page_config(page_title="📘 PDF RAG QA", layout="wide")
# Custom styles
st.markdown("""
<style>
.main {background-color: #f7faff;}
h1 {color: #4a4a8a;}
.stTextInput>div>div>input {border: 2px solid #d0d7ff;}
.stButton button {background-color: #4a4a8a; color: white;}
</style>
""", unsafe_allow_html=True)
st.title("📘 Ask Me Anything About Machine Learning")
st.caption("Using RAG (Retrieval-Augmented Generation) and a preloaded PDF")
# Load PDF from local file
PDF_FILE = "ml_large_dataset.pdf"
def load_pdf(file_path):
with open(file_path, "rb") as f:
reader = PyPDF2.PdfReader(f)
return [page.extract_text() for page in reader.pages]
def chunk_text(pages, max_len=1000):
text = " ".join(pages)
words = text.split()
return [' '.join(words[i:i+max_len]) for i in range(0, len(words), max_len)]
@st.cache_resource
def setup_rag():
pages = load_pdf(PDF_FILE)
chunks = chunk_text(pages)
model = SentenceTransformer('all-MiniLM-L6-v2')
embeddings = model.encode(chunks)
index = faiss.IndexFlatL2(embeddings.shape[1])
index.add(np.array(embeddings))
qa = pipeline("question-answering", model="deepset/roberta-base-squad2")
return chunks, model, index, qa
def retrieve_answer(question, chunks, model, index, qa_pipeline, k=6):
q_embed = model.encode([question])
_, I = index.search(np.array(q_embed), k)
context = "\n\n".join([chunks[i] for i in I[0]])
result = qa_pipeline(question=question, context=context)
return result['answer']
chunks, embed_model, faiss_index, qa_model = setup_rag()
st.subheader("💬 Ask a Question")
question = st.text_input("Enter your question:", placeholder="e.g., What is supervised learning?")
if question:
with st.spinner("🧠 Searching for the answer..."):
answer = retrieve_answer(question, chunks, embed_model, faiss_index, qa_model)
st.markdown("#### 📖 Answer:")
st.write(answer)
|