app.py

import streamlit as st
import json
from datasets import load_dataset

st.set_page_config(page_title="Kaggle Notebooks inspection", layout="wide")
st.title("Kaggle Notebooks inspection 🔍")

st.markdown("Here you can inspect Kaggle notebooks that were converted to python scripts and deduplicated.")
@st.cache()
def load_data():
    ds = load_dataset("loubnabnl/kaggle_scripts_dedup_subset", split="train")
    return ds

samples = load_data()
col1, _ = st.columns([2, 4])
with col1:
    index_example = st.number_input(f"Index of the chosen sample from the existing {len(samples)}", min_value=0, max_value=len(samples)-1, value=0, step=1)

st.code(samples[index_example]["script"])