Customizing validphys plots and other functionality

validphys comes with extensive capabilities for producing publication ready figures and other results. See for example How to plot PDFs, distances and luminosities or How to do a data theory comparison. Here we discuss what to do when these are not quite enough for a given application, for example when adding extremely specific plots or producing the final touches for plots for a publication or talk.

Tweaking Matplotlib styles

Some aspects of the appearance of the figure can be customized using Matplotlib stylesheets. The default styles we use can be found in the validphys.mplstyles module. A different style file can be passed using the --style flag of validphys (and other applications derived from reportengine). Note that the styles don’t compose and therefore it is advised to copy the default stylesheet and modify it as needed.

Note that several options haven’t been chosen at random. For example the first few entries in the color cycle are colorblind friendly and such that colors look OK when stacked with transparency on top of each other, for example for PDF plots.

Consider adding the functionality to master

New types of plots, as well as stylistic or functional enhancements to existing code can be added to validphys, following the appropriate process.

This option is strongly recommended and should be the default choice in most situations. While it requires some initial investment, in coming up with an appropriate The design of validphys 2 to make the required feature fit the rest of the system, as well meeting somewhat high coding standards, there are important benefits in exchange: The feature will get a few additional eyeballs and once merged it will be maintained and kept in sync with the rest of the code, making runcards using it much more likely to work in the future. That others can benefit from the work is of course also a good thing.

Hooking validphys to external code

In some situations the requirements for a given plot are rather esoteric and there is no way to add the functionality to the code economically. In such cases, external code can be used. Even so, consider upstreaming as much of the functionality as possible, to get the benefits discussed above.

There are two ways to take advantage of resources produced using the validphys execution model to process them further.

  • Using the API: It is possible to get some data from validphys using the validphys API and then use it in a script. This affords maximum flexibility, as the script can do anything. In exchange runcard based input processing or structured output folders aren’t readily available. Prefer this option for a very small project or when the task doesn’t fit the execution model of validphys for some reason.

  • Using extra modules: Additional Python modules or files can be passed to validphys using the --extra-modules (or -x) flag. The functions in these modules then act validphys providers and can take resources from validpys as input. This approach allows the immediate use of runcards or the default styles. One limitation is that there is currently no way of adding production rules or parsers in this way. Prefer this for actions that are too difficult to upstream to validphys, but should work as if they were internal. A minimal example for an external module could be:

    from matplotlib.figure import Figure
    from reportengine.figure import figure
    from validphys.commondataparser import load_commondata
    # A simple plot that probably should be in validphys to begin with.
    def plot_central_values(commondata):
        fig = Figure()
        ax = fig.subplots()
        return fig

    The action plot_central_values can now be used in a runcard:

        dataset: NMC
        - plot_central_values

Provided that validphys is invoked as validphys runcard.yaml -x

Note that both of these come at the cost of risking future breakage somewhat as we don’t guarantee any sort of stability on the internal interfaces.

Editing SVG files

SVG files store information on figures as sprites and text rather than pixels. These can then be edited with image editors such as Inkscape. It is possible to edit the text in the figure or change colors of individual lines. Note that this is the least maintainable approach as the modifications need to be applied manually every time the plot is updated. However it may be a good way to quickly enhance a plot for a presentation for example. To produce SVG files, pass the flag --formats svg when invoking validphys.