Pandas Exploration Toolkit
A toolkit to assist data exploration for both tabular and geospatial data. Inspired by pandas-profiling.
Planned Features for v1 Release:
- Validation against rules for Tabular data
- Visualizations of data
- Correlation analysis
Requirements
- Python 3.6
- geopandas>=0.4.0
Installation
git clone https://github.com/open-data-toronto/petk
cd petk
python setup.py install
Usage
import geopandas as gpd
import petk
df = gpd.read_file([data_path])
report = petk.DataReport(df)
# To display a high level description of the data
report.introduce
# To show a statistical breakdown of the data
report.describe()
# To validate the content of the data against fixed rules (currently only for GeoSpatial data)
report.validate(rules)
Where a sample of the rules could be
rules = {
'geometry': {
'get_invalids': True,
'get_slivers': {
'projection': 2019,
'area_thresh': 1,
'length_thresh': 1
}
}
}
Contribution
All contributions, bug reports, bug fixes, documentation improvements, enhancements and ideas are welcome.
Reporting issues
Please report issues here.
Contributing
Please develop in your own branch and create Pull Requests into the dev
branch when ready for review and merge.