Digautoprofiler NPM

Profile your Pandas Dataframes! Autoprofiler will automatically visualize your Pandas dataframes after every execution, no extra code necessary.

Autoprofiler allows you to spend less time specifying charts and more time interacting with your data by automatically showing you profiling information like:

Distribution of each column
Sample values
Summary statistics

Updates profiles as your data updates

screenshot of Autoprofiler

Autoprofiler reads your current Jupyter notebook and produces profiles for the Pandas Dataframes in your memory as they change.

https://user-images.githubusercontent.com/13400543/199877605-ba50f9c8-87e5-46c9-8207-1c6496bb3b18.mov

Install

To instally locally use pip and then open jupyter lab and the extension will be running.

pip install -U digautoprofiler

Please note, AutoProfiler only works in JupyterLab with version >=3.x, < 4.0.0.

Try it out

To try out Autoprofiler in a hosted notebook, use one of the options below

Jupyter Lite	Binder

Browser support: AutoProfiler has been developed and tested with Chrome.

Development Install

For development install instructions, see CONTRIBUTING.md.

If you're having install issues, see TROUBLESHOOTING.md.

Acknowledgements

Big thanks to the Rill Data team! Much of our profiler UI code is adapted from Rill Developer.

Citation

Please reference our VIS'23 paper:

@article{epperson23autoprofiler,
  title={Dead or Alive: Continuous Data Profiling for Interactive Data Science},
  author={Will Epperson and Vaishnavi Goranla and Dominik Moritz and Adam Perer},
  journal={IEEE Transactions on Visualization and Computer Graphics},
  year={2023},
  url={https://arxiv.org/abs/2308.03964}
}