Holoviz-based plotting in the lab - Primer¶

This notebook gives a brief intro to using holoviz-based plotting, incl our own tools that work with it.

To make sure interactivity works correctly, run this notebook in an environment that has all the right stuff installed (i.e., your labcore environment)

Some basic preliminaries¶

We assume that raw data is in DataDict formats. For processing and plotting we convert to either pandas.DataFrame (for ungridded data) or xarray.Dataset (for gridded data). This approach should be the general way we handle data in the lab.

Brief primer on vanilla holoviz¶

We have some tools on top of it for general easy data exploration.

But for data analysis it's probably very wise for everyone to learn how to use holoviz directly. It allows making custom plots extremely easily.

Docs are here: https://holoviz.org; see in particular holoviews and hvplot.

Here is a very simple example -- we use synthetic Rabi/Chevron data for that. But much more is possible -- check out the examples in the documentation for how to do different things.

In [6]:

# first, make a dataset in datadict
import numpy as np

from labcore.testing.dispersive_qubit_readout_data import chevron_dataset

raw_data = chevron_dataset(
    Omega_0=1e6,
    Delta_vals=np.linspace(-1e6, 1e6, 20),
    t_vals=np.linspace(0, 3e-6, 20),
    n=20,  # number of shots
)

type(raw_data), raw_data

# first, make a dataset in datadict import numpy as np from labcore.testing.dispersive_qubit_readout_data import chevron_dataset raw_data = chevron_dataset( Omega_0=1e6, Delta_vals=np.linspace(-1e6, 1e6, 20), t_vals=np.linspace(0, 3e-6, 20), n=20, # number of shots ) type(raw_data), raw_data

Out[6]:

(labcore.data.datadict.DataDict,
 signal: (400, 20)
   ⌙ repetition: (400, 20)
   ⌙ detuning (Hz): (400,)
   ⌙ time (s): (400,))

In [3]:

# next -- convert into xarray. since we know its a nice regular grid, that's the easiest choice here.
from labcore.data.datadict import datadict_to_meshgrid, dd2xr
from labcore.data.tools import split_complex

# with the holoviz (or all pydata) tools, complex data isn't always the best, so we split it immediately.
xrdata = split_complex(
    dd2xr(datadict_to_meshgrid(raw_data))
)
xrdata

# next -- convert into xarray. since we know its a nice regular grid, that's the easiest choice here. from labcore.data.datadict import datadict_to_meshgrid, dd2xr from labcore.data.tools import split_complex # with the holoviz (or all pydata) tools, complex data isn't always the best, so we split it immediately. xrdata = split_complex( dd2xr(datadict_to_meshgrid(raw_data)) ) xrdata

Out[3]:

<xarray.Dataset>
Dimensions:     (repetition: 20, detuning: 20, time: 20)
Coordinates:
  * repetition  (repetition) int64 1 2 3 4 5 6 7 8 9 ... 13 14 15 16 17 18 19 20
  * detuning    (detuning) float64 -1e+06 -8.947e+05 ... 8.947e+05 1e+06
  * time        (time) float64 0.0 1.579e-07 3.158e-07 ... 2.842e-06 3e-06
Data variables:
    signal_Re   (repetition, detuning, time) float64 2.085 1.837 ... 2.129 1.809
    signal_Im   (repetition, detuning, time) float64 0.4333 0.4132 ... 0.618

xarray.Dataset

• Dimensions:
  • repetition: 20
  • detuning: 20
  • time: 20
• Coordinates: (3)
  • repetition
    (repetition)
    int64
    1 2 3 4 5 6 7 ... 15 16 17 18 19 20

    array([ 1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
           19, 20])

  • detuning
    (detuning)
    float64
    -1e+06 -8.947e+05 ... 1e+06

    array([-1000000.      ,  -894736.842105,  -789473.684211,  -684210.526316,
            -578947.368421,  -473684.210526,  -368421.052632,  -263157.894737,
            -157894.736842,   -52631.578947,    52631.578947,   157894.736842,
             263157.894737,   368421.052632,   473684.210526,   578947.368421,
             684210.526316,   789473.684211,   894736.842105,  1000000.      ])

  • time
    (time)
    float64
    0.0 1.579e-07 ... 2.842e-06 3e-06

    array([0.000000e+00, 1.578947e-07, 3.157895e-07, 4.736842e-07, 6.315789e-07,
           7.894737e-07, 9.473684e-07, 1.105263e-06, 1.263158e-06, 1.421053e-06,
           1.578947e-06, 1.736842e-06, 1.894737e-06, 2.052632e-06, 2.210526e-06,
           2.368421e-06, 2.526316e-06, 2.684211e-06, 2.842105e-06, 3.000000e-06])

• Data variables: (2)
  • signal_Re
    (repetition, detuning, time)
    float64
    2.085 1.837 1.817 ... 2.129 1.809

    array([[[ 2.0851939 ,  1.83727863,  1.81690448, ...,  2.56526006,
              2.43875258,  2.58742792],
            [ 1.62842295,  1.55057649, -0.02121912, ...,  0.14267972,
              1.81500827,  2.38626988],
            [ 1.3066802 ,  2.09997738, -0.38328366, ...,  1.87326002,
              2.29177096,  1.68055914],
            ...,
            [ 2.65276905,  1.98888767,  0.2566441 , ..., -0.03979531,
             -0.07094792,  1.78484301],
            [ 2.00509194,  1.83970035,  1.70723588, ...,  0.09904967,
              1.27428081,  1.50922241],
            [ 2.71748561,  2.06437326, -0.06360528, ...,  1.57288186,
              1.73186957,  2.57015356]],
    
           [[ 1.9974292 ,  1.14630178, -0.37400418, ...,  1.9895205 ,
              2.24054055,  1.76566939],
            [ 2.04096079,  1.56035418,  0.08069875, ...,  2.01149441,
              3.30612649,  2.26998237],
            [ 1.26433686,  3.54597364, -0.34866773, ...,  0.12259455,
              0.79041058,  1.28952712],
    ...
            [ 2.30030213,  1.01658706,  1.71381811, ...,  0.99411573,
              2.16233766,  1.98668358],
            [ 1.74576817,  2.11499347,  2.13085457, ...,  2.13709   ,
              1.41863882,  2.97141426],
            [ 2.19827526,  0.21545599,  0.98735214, ...,  2.65736886,
              2.32386422,  2.20618771]],
    
           [[ 1.8289988 ,  2.59407019, -0.24254022, ...,  1.03664099,
              2.93782371,  2.21651199],
            [ 1.32646094,  2.07718757, -0.4232972 , ...,  1.29332442,
              2.06354481,  1.91395166],
            [ 2.61670771,  1.28755507,  2.04907833, ...,  0.29563388,
              1.67593694,  1.89946439],
            ...,
            [ 2.27551711,  2.6075095 ,  0.68144205, ...,  2.33161072,
             -0.50588016,  1.94095499],
            [ 1.92124748,  2.43519627,  0.45130262, ..., -0.11526138,
              2.62476728,  2.84275303],
            [ 2.30900089,  2.07135007,  1.68178912, ...,  0.19946094,
              2.1294448 ,  1.80945159]]])

  • signal_Im
    (repetition, detuning, time)
    float64
    0.4333 0.4132 ... 0.5651 0.618

    array([[[ 4.33331976e-01,  4.13180700e-01,  7.19167793e-01, ...,
             -7.55043555e-01,  5.64193310e-01, -1.97088202e-01],
            [-1.54007804e-02,  1.11123945e-02,  2.63470278e+00, ...,
              1.86262091e+00, -1.11053845e-01, -6.74806025e-01],
            [-6.97373221e-01, -1.09999182e+00,  1.97304894e+00, ...,
             -3.62671257e-01, -1.04984259e-01,  6.76603619e-01],
            ...,
            [ 2.72758276e-01, -6.68977378e-01,  1.36671376e+00, ...,
              1.64056199e+00,  2.86257172e+00, -2.38564185e-03],
            [-3.75613783e-01, -6.20009349e-01,  3.66205976e-01, ...,
              2.13175017e+00,  3.27313501e-01, -2.94930669e-01],
            [-7.22722183e-01, -3.09303221e-01,  1.63896237e+00, ...,
              3.78687500e-01, -2.91357835e-01,  3.11283098e-01]],
    
           [[-2.37387704e-01, -2.66721615e-01,  8.45059615e-01, ...,
              3.16332567e-01, -1.28795901e-01, -1.60166878e-01],
            [-2.25371161e-02, -3.53440534e-01,  2.56418621e+00, ...,
              8.63708698e-01,  5.64639703e-02,  4.28550084e-01],
            [ 7.63651604e-01,  5.60525287e-01,  2.26645695e+00, ...,
              1.97589030e+00,  2.03954829e+00, -2.02848155e-01],
    ...
            [ 8.96788128e-01,  2.55030593e-01, -2.91397072e-01, ...,
              3.06128869e-01,  6.63412542e-01,  5.38096369e-01],
            [-2.62282977e-01,  6.88560027e-01,  2.35201411e-01, ...,
              5.68684570e-02, -3.03798495e-01, -6.41625055e-01],
            [ 5.16818301e-01,  1.92932353e+00, -7.44890272e-01, ...,
             -7.87731533e-02,  4.77596691e-01, -1.67441382e-01]],
    
           [[ 4.60202794e-01,  1.72346126e-01,  1.73976574e+00, ...,
              1.58372237e-01,  9.49080071e-01,  4.20641983e-01],
            [ 5.46535526e-01,  2.26097247e-01,  1.87478713e+00, ...,
             -8.22224728e-01,  7.78867094e-01, -9.73560799e-01],
            [ 1.47087110e-01,  2.65841275e-02, -1.33910428e-01, ...,
              2.81750702e+00,  2.59548382e-01,  5.42427105e-01],
            ...,
            [ 9.80104391e-01, -4.86673487e-01,  1.75141841e+00, ...,
              5.58410981e-02,  1.88475913e+00, -3.76147790e-01],
            [ 3.20776538e-01, -7.45326801e-03,  1.78496994e+00, ...,
              2.10680104e+00, -5.79809980e-01, -8.44912031e-01],
            [-2.66867865e-01, -6.38558976e-01,  4.52101158e-01, ...,
              1.85990760e+00,  5.65132230e-01,  6.18006842e-01]]])

• Indexes: (3)
  • repetition
    PandasIndex

    PandasIndex(Index([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20], dtype='int64', name='repetition'))

  • detuning
    PandasIndex

    PandasIndex(Index([         -1000000.0,  -894736.8421052631,  -789473.6842105263,
            -684210.5263157894,  -578947.3684210526, -473684.21052631584,
             -368421.052631579, -263157.89473684214, -157894.73684210528,
            -52631.57894736843,   52631.57894736831,  157894.73684210517,
              263157.894736842,   368421.0526315789,   473684.2105263157,
             578947.3684210526,   684210.5263157894,   789473.6842105263,
             894736.8421052631,           1000000.0],
          dtype='float64', name='detuning'))

  • time
    PandasIndex

    PandasIndex(Index([                   0.0, 1.5789473684210527e-07, 3.1578947368421055e-07,
            4.736842105263158e-07,  6.315789473684211e-07,  7.894736842105264e-07,
            9.473684210526316e-07, 1.1052631578947369e-06, 1.2631578947368422e-06,
           1.4210526315789475e-06, 1.5789473684210528e-06,  1.736842105263158e-06,
           1.8947368421052632e-06, 2.0526315789473687e-06, 2.2105263157894738e-06,
           2.3684210526315793e-06, 2.5263157894736844e-06, 2.6842105263157895e-06,
            2.842105263157895e-06,                  3e-06],
          dtype='float64', name='time'))

• Attributes: (0)

In [4]:

# now we can directly use hvplot on this
import hvplot.xarray  # this import patches xarray objects so we can use hvplot on them.

# plot: average over repetition, then plot the real part as 2d map.
plot = xrdata.signal_Re.mean('repetition').hvplot.quadmesh(
    x='detuning',
    y='time',
)
plot

# now we can directly use hvplot on this import hvplot.xarray # this import patches xarray objects so we can use hvplot on them. # plot: average over repetition, then plot the real part as 2d map. plot = xrdata.signal_Re.mean('repetition').hvplot.quadmesh( x='detuning', y='time', ) plot

Out[4]:

In [ ]:

In [ ]:

In [ ]:

In [ ]:

In [5]:

# it's also pretty easy to make linecuts with sliders
_data = xrdata.mean('repetition')

# here: overlay line and scatter plot
# can change x to get a different cut
plot = _data.hvplot.line(x='time') * _data.hvplot.scatter(x='time')
plot

# it's also pretty easy to make linecuts with sliders _data = xrdata.mean('repetition') # here: overlay line and scatter plot # can change x to get a different cut plot = _data.hvplot.line(x='time') * _data.hvplot.scatter(x='time') plot

Out[5]:

In [7]:

# looking at readout histograms -- use simply all points in the set
# hexbin is a very simple way to have a quick look at this. 
# doesn't offer too much control, but is 
xrdata.hvplot(
    kind='hexbin',
    aspect=1,
    groupby=[],
    x='signal_Re',
    y='signal_Im',
)

# looking at readout histograms -- use simply all points in the set # hexbin is a very simple way to have a quick look at this. # doesn't offer too much control, but is xrdata.hvplot( kind='hexbin', aspect=1, groupby=[], x='signal_Re', y='signal_Im', )

Out[7]:

In [8]:

# if we want to select based on some time/detuning values...
xrdata.hvplot(
    kind='hexbin',
    aspect=1,
    groupby=['time', 'detuning'],
    x='signal_Re',
    y='signal_Im',
    xlim=(-3,3),
    ylim=(-3,3)
)

# if we want to select based on some time/detuning values... xrdata.hvplot( kind='hexbin', aspect=1, groupby=['time', 'detuning'], x='signal_Re', y='signal_Im', xlim=(-3,3), ylim=(-3,3) )

Out[8]:

In [ ]:

In [ ]:

In [ ]:

In [ ]: