app.py

import streamlit as st
from pydantic_ai import Agent
from pydantic_ai.models.groq import GroqModel
import nest_asyncio
import pdfplumber
import os

api_key = os.getenv("API_KEY")
data = []

#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

model = GroqModel('llama-3.1-70b-versatile', api_key = api_key)

def extract_data(feed):
    
    with pdfplumber.load(feed) as pdf:
        pages = pdf.pages
        for p in pages:
            data.append(p.extract_tables())
    return None 

uploaded_file = st.file_uploader('Choose your .pdf file', type="pdf")
if uploaded_file is not None:
    df = extract_data(uploaded_file)

if data is not None:
    st.caption(data)