GWpy is a large package with a number sub-packages, so importing the root package via:
>>> import gwpy
isn’t going to be very useful. Instead, it is best to import the classes you need directly, for example:
>>> from gwpy.timeseries import TimeSeries
GWpy is designed to be an object-oriented programming package, that is, data objects are the central focus of the package. Each data object is represented as an instance of a class, describing its properties and the data it holds.
In order to generate a new instance of any class, you should use the standard constructor, or any of the classmethod functions. For example, a TimeSeries can be generated from an existing data array:
>>> from gwpy.timeseries import TimeSeries
>>> mydata = TimeSeries([1,2,3,4,5,6,7,8,9,10], sample_rate=1, epoch=0)
or by downloading it from the relevant network data server:
>>> from gwpy.timeseries import TimeSeries
>>> mydata = TimeSeries.fetch('H1:LDAS-STRAIN', 964656015, 964656615)
From there, anything you might want to do with the TimeSeries can be done directly from the object itself, rather than passing it into a separate function. For example, if you want to calculate the amplitude spectral density of that series:
>>> spectrum = mydata.asd(4)
where the 4 argument tells the asd() method to generate a Welch average ASD FrequencySeries (by default) with non-overlapping, 4-second Fourier transforms.
Then you have a new object, a FrequencySeries, with its own methods and properties.
There are a small number of core objects provided by GWpy, each representing the standard data products of a gravitational-wave interferometer, or their derivatives. These are
TimeSeries |
A time-domain data array |
FrequencySeries |
A data array holding some metadata to represent a frequency series |
Spectrogram |
A 2D array holding a spectrogram of time-frequency data |
DataQualityFlag |
A representation of a named set of segments. |
The following pages in this documentation give full descriptions of how to read and manipulate data, and access data segments, amongst other things. The remainder of this page outlines a few more key concepts surrounding these core objects.
Each of these objects comes with a standard input method read(), which will accept any of a set of registered file formats for the respective class.
For example, you can read a TimeSeries from a GWF-format file as follows:
>>> from gwpy.timeseries import TimeSeries
>>> data = TimeSeries.read('/archive/frames/A6/L0/LLO/L-R-10670/L-R-1067042880-32.gwf', 'L1:PSL-ODC_CHANNEL_OUT_DQ')
Similary, each class has a standard output method write(), again accepting a number of recognised formats.
Analogous to the unified input/output system, each of the standard objects comes with a plot() method, display that object on a figure using the matplotlib display library.
Following from the above example, the TimeSeries data can be displayed via:
>>> plot = data.plot()
If you have an interactive backend, you can immediately show the figure on your screen via:
>>> plot.show()