Live plotting

add_subplots

As we’ve seen so far, if you just want to plot a Parameter in each of the subplots of the live plotting window, you can simply pass Parameters to loop.run(). You can also pass them when you define the Loop using the plot argument, e.g.:

loop=qc.loop1d(sweep_parameter=qdac.ch01.volt,
           start=0,stop=1,num=101,delay=0.03,
           device_info='Device1',
           instrument_info=f'ACdiv=1e5 DCdiv=1e3 freq={li.osc0_freq():.6g} Hz',
           measure=[currentX, currentY, voltageX, voltageY, resistance],
           plot=[currentX,voltageX,resistance])

For more control over live plotting options, you can use the qc.live_plot() function which returns a qc.Plot object.

pp=qc.live_plot()

You can optionally initialise it with parameters to plot, as in loop.run().

pp=qc.live_plot(currentX,voltageX,resistance)

or equivalently, add them afterwards using add_subplots()

pp=qc.live_plot()
pp.add_subplots(currentX,voltageX,resistance)

What is being plotted?

Each DataSetPP consists of a number of DataArray elements, which hold the stored data for each Parameter. What you are actually plotting is data from these DataArray elements. When you declare your loop, a summary of the DataSetPP is printed, showing the array_id for each DataArray and its shape. You can access the DataSetPP for a particular loop via loop.data_set, and the individual DataArray elements via e.g. loop.data_set.param1 or loop.data_set.arrays['param1'].

This is important to understand for the next bit!

add

The add method allows the most control over subplots.

pp=qc.live_plot()
pp.add(param1, title='Parameter 1', name='Parameter 1', subplot=0)
pp.add(param2, title='Parameter 2', name='Parameter 2', subplot=1)

The first argument to add can be:

A DataArray object, such as loop.data_set.param1.
An array_id string, e.g. 'param1'.
A Parameter object, if the DataSetPP has only one DataArray per Parameter.

Method 1 always works. Method 2 will almost always work, and Method 3 can work in simple cases. Methods 2 and 3 work by looking for the array_id or Parameter name in the most recently created DataSetPP, and then adding the relevant DataArray. This means these methods only work if the DataSetPP of interest was indeed the last created DataSetPP. Otherwise, you can supply the data_set argument to specify which DataSetPP to use.

pp.add('example_array_id', title='Parameter 1', name='Parameter 1', subplot=2, data_set=old_loop.data_set)

After you have set up the plot as you prefer, you can run the loop without any arguments, i.e. data=loop.run().

Some tricks

To add multiple elements to the same subplot, simply use the same index for subplot=.

pp.add(param1, title='Parameter 1', name='Parameter 1', subplot=0)
pp.add(param2, title='Parameter 2', name='Parameter 2', subplot=0)

By default, the provided dataset element is plotted as the y-axis for 1D plots and the z-axis for 2D plots, with the independent variables plotted on their respective axes. You can also specify the x-axis (and y-axis for 2D plots) explicitly by providing the appropriate number of dataset elements as arguments, e.g.:

pp.add(data.paramx, data.paramy, data.paramz, title='Parameter 1', name='Parameter 1', subplot=0)
pp.add('paramx2','paramy2','paramz2', title='Parameter 2', name='Parameter 2', subplot=1)

You can of course add data from a different DataSetPP. This is most commonly desired if you want to run a new experiment and plot it alongside a previous experiment’s data. Simply define a new loop, but not a new plot window.

loop=qc.loop1d(sweep_parameter=qdac.ch01.volt,
           start=0,stop=1,num=101,delay=0.03,
           device_info='Device1',
           instrument_info=f'ACdiv=1e5 DCdiv=1e3 freq={li.osc0_freq():.6g} Hz',
           measure=[param1, param2]
pp.add(param1, title='Parameter 1', name='Parameter 1', subplot=2)
new_loop.run()

Back to add_subplots

Similarly to add, add_subplots() will also accept any of DataArray, array_id or Parameter. So far we just provided Parameter because it’s usually easiest, but in some situations you may want to use DataArray or array_id. For example, in loop2dUD, each Parameter has two corresponding DataArray elements. While add doesn’t know which one to plot, and therefore won’t let you supply a Parameter at all, add_subplots will do the opposite; it will plot both DataArray elements. If you only want to plot one, you can specify e.g. using the array_id:

pp.add_subplots('currentX_1','voltageX_2')

In some cases it is necessary to supply either the DataArray or the array_id; one example is for MultiParameters where each component of the MultiParameter generates a corresponding DataArray. But since a component of a MultiParameter is not itself a Parameter, you must of course use either DataArray or array_id.

Finally, you can mix and match as you please. The following will work just fine:

pp.add_subplots('currentX_1',voltageX,loop.data_set.resistance)

Publisher

In the back end, live_plot works by setting the publisher attribute of the DataSetPP to the Plot object. The publisher method sends the data to the plot from memory, bypassing saving to disk; part of the reason live_plot is so fast. When you use add or add_subplots, qcodes++ automatically assigns the publisher of the most recently defined dataset to the Plot you are adding to. Therefore, as long as you use add or add_subplots after creating the DataSetPP you want to add, and before creating a new DataSetPP, the publisher is set automatically. This covers 99% of cases. In the case where you cannot do things in this order, use e.g. data.publisher=pp to set the publisher manually.