streamlit-excel 0.0.2

Build Excel-style filter widgets for large pandas DataFrames in Streamlit

Welcome to the streamlit-excel package!

This package provides a Streamlit component for filtering large pandas DataFrames (e.g., over 100,000 rows) through an intuitive, Excel-like interface. It enables interactive filtering directly within a Streamlit app, allowing users to explore and refine data visually and efficiently.

Designed for simplicity and seamless integration, the component is ideal for data scientists and developers building interactive data visualizations.

Key Features

• Easy Integration: Initialize the component with a Pandas DataFrame using the Table class and store it in the session state.
• Flexible Filtering: Use Table.show_filter_widget to render a filter widget for specific columns. The widget can be placed anywhere in the app, including the sidebar.
• Dynamic Views: Access the filtered DataFrame via the Table.view property.
• Multiple Tables: Support for multiple tables by creating separate Table instances with unique key parameters.
• Custom Column Names: Customize display names in the filter widget by passing a dictionary to the mapper parameter when creating the Table instance.

Installation

$ pip install streamlit-excel streamlit pandas

Usage

Below is a simple example demonstrating how to use the streamlit-excel package to create a filterable table in a Streamlit app. In this example, we generate a random dataset with 100,000 rows and 5 columns to showcase the package's scalability.

Save the following code to app.py:

import streamlit as st
import streamlit_excel
import pandas as pd
import numpy as np
import random
from datetime import datetime

if "example_table" not in st.session_state:
    n = 100000
    df = pd.DataFrame({
        "Name": np.random.choice(["Emily Johnson", "Michael Smith", "Lena Corwin", "Talia Vexley", "Kai Renford"], n),
        "City": np.random.choice(["Kyoto", "Cape Town", "Vancouver", "New York City", "Paris"], n),
        "Color": np.random.choice(["Red", "Green", "Yellow", "Blue"], n),
        "Start": [datetime(random.randint(1999, 2022), random.randint(1, 12), random.randint(1, 28)) for _ in range(n)],
        "End": [datetime(random.randint(2023, 2025), random.randint(1, 12), random.randint(1, 28)) for _ in range(n)],
    })
    st.session_state.example_table = streamlit_excel.Table(df, "example_table")

st.session_state.example_table.show_filter_widget("Example Table", ["Name", "City", "Color", "Start", "End"])
st.dataframe(st.session_state.example_table.view)
st.dataframe(st.session_state.example_table.view.describe(include="all"))

Run the app:

$ streamlit run app.py