**dgpost**: datagram post-processing toolkit
============================================
.. image:: https://badgen.net/badge/docs/dgbowl.github.io/grey?icon=firefox
:target: https://dgbowl.github.io/dgpost
.. image:: https://badgen.net/pypi/v/dgpost/?icon=pypi
:target: https://pypi.org/project/dgpost
.. image:: https://badgen.net/github/tag/dgbowl/dgpost/?icon=github
:target: https://github.com/dgbowl/dgpost
A set of datagram postprocessing tools, functions, visualisations and export scripts.
dgpost is a Python library for reproducible postprocessing of time-resolved data.
It provides functions to load data from `datagrams` and :class:`pd.DataFrames`,
automatic datapoint interpolation using unix timestamps, a library of transformation
functions with implementations of common calculations in chemistry, catalysis,
electrochemistry, etc., and convenience functions to save and plot the calculated data.
dgpost relies on the `dgpost-recipe` schema, defined in the
:mod:`dgbowl_schemas.dgpost_recipe` module of that package.
In addition to the above, dgpost is fully unit- and uncertainty-aware, using the
`pint `_ and
`uncertainties `_ packages.
.. toctree::
:maxdepth: 3
:caption: dgpost user manual
:hidden:
installation
usage
features
version
.. toctree::
:maxdepth: 2
:caption: dgpost function library
:hidden:
apidoc/dgpost.transform.catalysis
apidoc/dgpost.transform.chromatography
apidoc/dgpost.transform.electrochemistry
apidoc/dgpost.transform.impedance
apidoc/dgpost.transform.rates
apidoc/dgpost.transform.reflection
apidoc/dgpost.transform.table
.. toctree::
:maxdepth: 2
:caption: dgpost utility library
:hidden:
apidoc/dgpost.utils.load
apidoc/dgpost.utils.extract
apidoc/dgpost.utils.transform
apidoc/dgpost.utils.plot
apidoc/dgpost.utils.save
.. toctree::
:maxdepth: 1
:caption: dgpost developer manual
:hidden:
contributing
apidoc/dgpost