File size: 3,519 Bytes
db2f74d 0adcf3b 044b536 db2f74d 044b536 db2f74d 0adcf3b 044b536 db2f74d 044b536 db2f74d 00fa616 db2f74d 00fa616 db2f74d 00fa616 db2f74d |
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 |
# app.py
import streamlit as st
from langchain.document_loaders import PyPDFLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.embeddings import SentenceTransformerEmbeddings
from langchain.vectorstores import FAISS
import os
from groq import Groq
from io import BytesIO
import tempfile
# Load PDF (Corrected)
def load_pdf(uploaded_file):
try:
bytes_data = uploaded_file.getvalue()
with tempfile.NamedTemporaryFile(delete=False, suffix=".pdf") as temp_file:
temp_file.write(bytes_data)
temp_file_path = temp_file.name
loader = PyPDFLoader(temp_file_path)
documents = loader.load()
os.remove(temp_file_path) # Clean up
return documents
except Exception as e:
st.error(f"Error loading PDF: {e}")
return None
# Chunking (with error handling)
def chunk_text(documents):
try:
text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
chunks = text_splitter.split_documents(documents)
return chunks
except Exception as e:
st.error(f"Error chunking text: {e}")
return None
# Embeddings and Vectorstore (with error handling)
def create_embeddings_and_store(chunks):
try:
embeddings = SentenceTransformerEmbeddings(model_name="all-mpnet-base-v2") # Or other suitable model
db = FAISS.from_documents(chunks, embeddings)
return db
except Exception as e:
st.error(f"Error creating embeddings: {e}")
return None
# Groq interaction (with more robust error handling and correct secret access)
def query_groq(query, db):
try:
groq_api_key = st.secrets["GROQ_API_KEY"] # Use st.secrets for Hugging Face secrets
docs = db.similarity_search(query) # Similarity search
context = "\n".join([doc.page_content for doc in docs])
client = Groq(api_key=groq_api_key) # Pass the secret to the client
prompt = f"""Use the following context to answer the question: {query}\n\nContext:\n{context}"""
chat_completion = client.chat.completions.create(
messages=[{"role": "user", "content": prompt}],
model="llama-3.3-70b-versatile", # Or other suitable open-source model compatible with Groq
)
return chat_completion.choices[0].message.content
except Exception as e:
st.error(f"Error querying Groq: {e}")
return None
# Streamlit app
st.title("RAG Application")
uploaded_file = st.file_uploader("Upload PDF", type="pdf")
if uploaded_file is not None:
with st.spinner("Processing PDF..."):
documents = load_pdf(uploaded_file)
if documents: # Check if PDF loaded successfully
chunks = chunk_text(documents)
if chunks: # Check if chunks were created successfully
db = create_embeddings_and_store(chunks)
if db: # Check if embeddings were created successfully
st.success("PDF processed!")
query = st.text_area("Enter your query")
if st.button("Submit"):
if query:
with st.spinner("Querying..."):
answer = query_groq(query, db)
if answer: # Check if query was successful
st.write(answer)
else:
st.warning("Please enter a query.") |