// the find
adamerose/PandasGUI
A GUI for Pandas DataFrames
PandasGUI is a desktop GUI (PyQt-based) for inspecting, filtering, and plotting Pandas DataFrames without writing throwaway `print`/`matplotlib` code. It's aimed at people doing exploratory data analysis who want a spreadsheet-like view and quick interactive charts on top of DataFrames they already have in a Python session.
The API is genuinely one line (`show(df)`) and handles multiple DataFrames and MultiIndex columns, which is more than most pandas viewers bother with. Interactive plotting is wired through `jotly.py` (a Plotly layer) rather than static matplotlib, so you get zoom/pan/hover for free. It covers real workflow gaps pandas itself lacks in a terminal — drag-and-drop CSV import, a filter builder, and copy/paste editing — and at 3.2k stars it's clearly been used enough to sand down the obvious rough edges.
It's still 0.x with the README explicitly warning of breaking changes, and the last push was May 2025 with no visible CI (no `.github/workflows`, just issue templates), so you can't tell if `master` is actually green. It's a hard dependency on Qt, meaning no headless/server use and a heavier install than the problem calls for. `show()` downloads sample datasets on first use when you just import the package's dataset helpers, which is an easy source of a surprising network call in a restricted environment. Being a GUI, it's also inherently unscriptable/non-reproducible — fine for ad hoc exploration, useless for anything you want to automate or put in a notebook pipeline.