selasviz

An interactive 2D visualization workbench for Pandas DataFrames, built with Panel and HoloViews. It supports scatter, hexbin, and datashader views.

This project is managed with uv.

Installation

1) Add to a project

With uv:

uv add selasviz

With pip:

pip install selasviz

After installing into your environment, the CLI entry selasviz is also available.

2) Use as a tool (CLI only)

Use this when you only need the CLI and do not want to add selasviz as a project dependency.

Run directly with uvx (temporary installation):

uvx --from selasviz selasviz --help

Or install as a global tool with uv:

uv tool install selasviz

Then run:

selasviz --help

Quick Start

import pandas as pd
from selasviz import launch_explorer

# Load your data
df = pd.read_csv("your_data.csv")

# Launch and serve the explorer
launch_explorer(df, port=5006, show=True)

CLI (FITS)

The CLI currently only supports FITS table files.

Launch directly from a FITS table file:

selasviz data.fits

Optional arguments:

selasviz data.fits --title "Data Explorer" --port 5006

Development (uv)

git clone https://github.com/Straw674/selasviz
cd selasviz
uv sync --extra dev

Run the standalone demo script from a source checkout:

uv run python scripts/demo_explorer.py --rows 50000 --title "Demo Explorer"

Release Versioning

This repository uses Semantic Versioning with python-semantic-release. The next version is inferred from Conventional Commit prefixes:

• fix: -> patch bump
• feat: -> minor bump
• feat!: or BREAKING CHANGE: -> major bump

Run locally:

uv sync --extra dev
uv run semantic-release version

The GitHub workflow also runs this automatically on pushes to main.

License

MIT License