dgpost 2.0a4

datagram post-processing toolkit

dgpost: datagram post-processing toolkit

Set of tools to post-process raw instrument data in yadg's datagram format, and tabulated data imported into pd.DataFrames.

Capabilities:

dgpost is indended to be used as part of your data processing pipeline, and works best with a series of timestamped data.

Write a recipe in yaml, and post-process your data from yadg.datagrams or pd.DataFrames in a reproducible fashion, while keeping provenance information, and without touching the original data files.

Post-process your data into pre-defined figures for your reports, or simply export your collated pd.DataFrame into one of the several supported formats!

Use dgpost in your Jupyter notebooks by importing it as a python package: import dgpost.utils to access the top-level functions for loading, extracting and exporting data; or import dgpost.transform to access the library of validated transform functions.

Features:

dgpost can load data from multiple file formats, extract data from those files into pd.DataFrames and automatically interpolate the datapoints along the time-axis (generally the index of the pd.DataFrame) as necessary, transform the created tables using functions from the built-in library, plot data from those tables using its matplotlib interface, and save the tables into several output formats.

Of course, dgpost is fully unit-aware, and supports values with uncertainties by using the pint.Quantity and uncertainties.ufloat under the hood.

For a further overview of features, see the project documentation.

Contributors:

• Peter Kraus
• Ulrich Sauter