This library is named for the wonderful coffee shop Tommy in Montreal where, as a good friend said: "the plants really tie the room together." It is my hope that this library will tie together pandas, sklearn, and matplotlib in ways that are useful for the practicing data scientist.
indoorplants provides data analysis and model validation tools, and the top-level namespace is broken out as such. Most of the code assumes that data is stored in DataFrame instances, and models are assumed to provide the sklearn interface. This code is built for Python 3.6+.
On the analysis side of things:
analysis.exploratoryaids with exploratory data analysisanalysis.wrangleis for data wrangling, including the detection of leaky features in classification problemsanalysis.featuresat the moment allows for calculating feature distances
On the validation side of things:
validation.crossvalidateprovides the core cross-validation functionsvalidation.curvesoffers curves for ML model evaluation, including: validation, learning, and calibration
Example usage can be found in the census_data_analysis notebook.
You can run pip install indoorplants to get this repo.