We're thrilled to feature Mike Kushman in this edition of the Preswald Community Showcase! 🎉 Mike has built an impressive Preswald app that analyzes and visualizes wine quality data using interactive charts and tables.
🌟 Project Overview
📊 Preswald and Analytics
This blog explores how Preswald utilizes analytics to evaluate wine quality, providing insights into chemical compositions, rating trends, and production attributes through dynamic visualizations.
Project Name: Wine Quality Explorer
Built by: Mike Kushman
Dataset: Wine quality dataset
Check out the project on GitHub
📊 Description
MJ's project provides an interactive dashboard for exploring wine quality data. The app allows users to:
- Filter wines by quality score, acidity, and alcohol content.
- Visualize trends using Plotly charts to analyze chemical composition and rating distributions.
- Compare wine attributes dynamically in an interactive table.
- Gain insights into factors affecting wine quality and taste profiles.
With Preswald, MJ has created a seamless user experience, where users can dynamically adjust inputs and get instant updates without writing frontend code.
🚀 Code Snippets
Here’s a sneak peek at how MJ's project is structured:
Loading the Data
from preswald import connect, get_df, text
connect() # Establish connection to data
# Load wine quality dataset
wine_df = get_df("wine_quality")
Creating an Interactive Quality Distribution Chart
from preswald import plotly
import plotly.express as px
fig = px.histogram(wine_df, x='quality', nbins=10, title='Wine Quality Distribution')
plotly(fig)
Displaying Wine Quality Data in a Table
from preswald import table
table(wine_df[['wine_type', 'quality', 'alcohol', 'acidity']])
With just a few lines of Python, MJ has built a fully interactive data app powered by Preswald! 🚀
What is Preswald?
Preswald is an open-source framework for building data apps, dashboards, and internal tools with just Python. It provides pre-built UI components like tables, charts, and forms, so you don't have to write frontend code. Users can interact with your app, changing inputs, running queries, and updating visualizations, without you needing to manage the UI manually.
Preswald tracks state and dependencies, making computations efficient by updating only when necessary. It uses a workflow DAG to manage execution order, ensuring performance and predictability. Preswald allows you to turn Python scripts into shareable, production-ready applications easily.
Key Features
- Add UI components to Python scripts: user-interactive buttons, text inputs, tables, and charts.
- Stateful execution with automatic state tracking and updates.
- Structured computation using a DAG-based execution model.
- Deploy with a single command.
- Query and display live data from various sources.
- Build interactive reports and dashboards.
- Easy local or cloud hosting.
- Shareable via a simple link.
🚀 Getting Started
Installation
First, install Preswald via pip: https://pypi.org/project/preswald/
pip install preswald
👩💻 Quick Start
1. Initialize a New Project
To start using Preswald, initialize a new project:
preswald init my_project
cd my_project
This creates a folder my_project with essential files:
hello.py: Your first Preswald app.preswald.toml: Settings for your app.secrets.toml: Secure sensitive information..gitignore: Keepsecrets.tomlsafe from version control.
2. Write Your First App
Open hello.py and write:
from preswald import text, plotly, connect, get_df, table
import pandas as pd
import plotly.express as px
text("# Welcome to Preswald!")
text("This is your first app. 🎉")
# Load CSV data
connect() # loads default sample CSV
df = get_df('sample_csv')
# Create a scatter plot
fig = px.scatter(df, x='quantity', y='value', text='item',
title='Quantity vs. Value')
fig.update_traces(textposition='top center')
fig.update_layout(template='plotly_white')
# Show plot and data
table(df)
3. Run Your App
Launch the app with:
preswald run
4. Deploy Your App to the Cloud
Deploy your app using:
preswald deploy --target structured
Your app will be built and accessible online.
Huge Thanks to Mike Kushman!
A big thank you to Mike Kushman for sharing their work and inspiring the Preswald community! Want to see more of their work? Check them out on GitHub | LinkedIn.
Want to Contribute?
Got a cool idea for a Preswald app? We'd love to see it! Get started here: https://github.com/StructuredLabs/preswald and get featured in our next Community Showcase.
Happy building! 🚀
