wwdata 0.2.0

Data analysis package aimed at data obtained in the context of (waste)water

wwdata

Data analysis package aimed at data obtained in the context of (waste)water

• Free software: GNU General Public License v3

• Documentation: https://ugentbiomath.github.io/wwdata-docs/

• Funding: Waterboard De Dommel

• Context: PhD research at BIOMATH, Ghent University

Structure

The package contains one class and three subclasses, all in separate .py files. Division in subclasses is based on the type of data:

• online data from full scale installations (OnlineSensorBased)

• online data from lab experiments (LabSensorBased)

• offline data obtained from lab experiments (LabExperimentBased).

Jupyter notbeook files (.ipynb) illustrate the use of the available functions. The most developed class is the OnlineSensorBased one. The workflow of this class is shown in below Figure, where OSB represents an OnlineSensorBased object. Main premises are to never delete data but to tag it and to be able to check the reliability when gaps in datasets are filled.

Examples

For the workflow with code and more specific examples, check out the Showcase Jupyter Notebook(s) included as documentation of the package.

Credits

This package was created with support from Cookiecutter and the audreyr/cookiecutter-pypackage project template, as well as this GitHub page, provided by Daler and explaining how to use sphinx documentation generation in combination with GitHub Pages.

History

0.1.0 (2017-10-23)

First release on PyPI.

The wwdata (wastewater data) package is meant to make data analysis, validation and filling of data gaps more streamlined. It contains code to do all this, while also providing simple visualisations of the whole procedure.

The package was (and is) developed in the framework of PhD research, involving the modelling of a full scale wastewater treatment plant (WWTP). Online measurements at the plant are available, but as with all data, is not perfect and therefor needs validation. The gap filling originated from the need to have high-frequency influent data available to run the WWTP model with.

0.2.0 (2018-06-12)

Second release on PyPI.

The wwdata (wastewater data) package is meant to make data analysis, validation and filling of data gaps more streamlined. It contains code to do all this, while also providing simple visualisations of the whole procedure.

The package was (and is) developed in the framework of PhD research, involving the modelling of a full scale wastewater treatment plant (WWTP). Online measurements at the plant are available, but as with all data, is not perfect and therefor needs validation. The gap filling originated from the need to have high-frequency influent data available to run the WWTP model with.

New in version 0.2.0:

• Bug fixes

• Addition of an only_checked argument to multiple functions to allow application of the function to only the validated data points (‘original’ in self.meta_valid).

• Extended, improved and customized documentation website (generated with sphinx).

• Extended and improved Jupyter Notebook for documentation.

• Improved visualisation for get_correlation: a prediction band based on the obtained correlation is now included in the produced scatter plot.

Known bugs:

• See (open issues on Github)[https://github.com/UGentBiomath/wwdata/issues])