class Driver(LabberDriver):
"""This class implements a multi-qubit pulse generator."""
def performOpen(self, options={}):
"""Perform the operation of opening the instrument connection."""
# init variables
self.sequence = None
self.sequence_to_waveforms = SequenceToWaveforms()
self.waveforms = {}
# always create a sequence at startup
name = self.getValue('Sequence')
self.sendValueToOther('Sequence', name)
def performSetValue(self, quant, value, sweepRate=0.0, options={}):
"""Perform the Set Value instrument operation."""
# only do something here if changing the sequence type
if quant.name == 'Sequence':
# create new sequence if sequence type changed
new_type = SEQUENCES[value]
if not isinstance(self.sequence, new_type):
# create a new sequence object
if value == 'Custom':
# for custom python files
path = self.getValue('Custom Python file')
(path, modName) = os.path.split(path)
sys.path.append(path)
modName = modName.split('.py')[0] # strip suffix
mod = importlib.import_module(modName)
# the custom sequence class has to be named
# 'CustomSequence'
if not isinstance(self.sequence, mod.CustomSequence):
self.sequence = mod.CustomSequence()
else:
# standard built-in sequence
self.sequence = new_type()
elif (quant.name == 'Custom Python file' and
self.getValue('Sequence') == 'Custom'):
# for custom python files
path = self.getValue('Custom Python file')
(path, modName) = os.path.split(path)
modName = modName.split('.py')[0] # strip suffix
sys.path.append(path)
mod = importlib.import_module(modName)
# the custom sequence class has to be named 'CustomSequence'
if not isinstance(self.sequence, mod.CustomSequence):
self.sequence = mod.CustomSequence()
return value
def performGetValue(self, quant, options={}):
"""Perform the Get Value instrument operation."""
# ignore if no sequence
if self.sequence is None:
return quant.getValue()
# check type of quantity
if (quant.name.startswith('Voltage, QB') or
quant.name.startswith('Single-shot, QB')):
# perform demodulation, check if config is updated
if self.isConfigUpdated():
# update sequence object with current driver configuation
config = self.instrCfg.getValuesDict()
self.sequence.set_parameters(config)
self.sequence_to_waveforms.set_parameters(config)
# get qubit index and waveforms
n = int(quant.name.split(', QB')[1]) - 1
demod_iq = self.getValue('Demodulation - IQ')
if demod_iq:
signal_i = self.getValue('Demodulation - Input I')
signal_q = self.getValue('Demodulation - Input Q')
else:
signal = self.getValue('Demodulation - Input')
ref = self.getValue('Demodulation - Reference')
# perform demodulation
if demod_iq:
value = self.sequence_to_waveforms.readout.demodulate_iq(
n, signal_i, signal_q, ref)
else:
value = self.sequence_to_waveforms.readout.demodulate(
n, signal, ref)
# average values if not single-shot
if not quant.name.startswith('Single-shot, QB'):
value = np.mean(value)
elif quant.isVector():
# traces, check if waveform needs to be re-calculated
if self.isConfigUpdated():
# update sequence object with current driver configuation
config = self.instrCfg.getValuesDict()
self.sequence.set_parameters(config)
self.sequence_to_waveforms.set_parameters(config)
# check if calculating multiple sequences, for randomization
if config.get('Output multiple sequences', False):
# create multiple randomizations, store in memory
n_call = int(config.get('Number of multiple sequences', 1))
calls = []
for n in range(n_call):
config['Randomize'] += 1
calls.append(
copy.deepcopy(
self.sequence_to_waveforms.get_waveforms(
self.sequence.get_sequence(config))))
# after all calls are done, convert output to matrix form
self.waveforms = dict()
n_qubit = self.sequence.n_qubit
# Align RB waveforms to end
align_RB_to_end = config.get('Align RB waveforms to end', False)
# start with xy, z and gate waveforms, list of data
for key in ['xy', 'z', 'gate']:
# get size of longest waveform
self.waveforms[key] = []
for n in range(n_qubit):
length = max([len(call[key][n]) for call in calls])
# build matrix
datatype = calls[0][key][n].dtype
data = np.zeros((n_call, length), dtype=datatype)
for m, call in enumerate(calls):
if align_RB_to_end:
data[m][-len(call[key][n]):] = call[key][n]
else:
data[m][:len(call[key][n])] = call[key][n]
self.waveforms[key].append(data)
# same for readout waveforms
for key in ['readout_trig', 'readout_iq']:
length = max([len(call[key]) for call in calls])
datatype = calls[0][key].dtype
data = np.zeros((n_call, length), dtype=datatype)
for m, call in enumerate(calls):
if align_RB_to_end:
data[m][-len(call[key]):] = call[key]
else:
data[m][:len(call[key])] = call[key]
self.waveforms[key] = data
else:
# normal operation, calcluate waveforms
self.waveforms = self.sequence_to_waveforms.get_waveforms(
self.sequence.get_sequence(config))
# get correct data from waveforms stored in memory
value = self.getWaveformFromMemory(quant)
else:
# for all other cases, do nothing
value = quant.getValue()
return value
def getWaveformFromMemory(self, quant):
"""Return data from already calculated waveforms."""
# check which data to return
if quant.name[-1] in ('1', '2', '3', '4', '5', '6', '7', '8', '9'):
# get name and number of qubit waveform asked for
name = quant.name[:-1]
n = int(quant.name[-1]) - 1
# get correct vector
if name == 'Trace - I':
if self.getValue('Swap IQ'):
value = self.waveforms['xy'][n].imag
else:
value = self.waveforms['xy'][n].real
elif name == 'Trace - Q':
if self.getValue('Swap IQ'):
value = self.waveforms['xy'][n].real
else:
value = self.waveforms['xy'][n].imag
elif name == 'Trace - Z':
value = self.waveforms['z'][n]
elif name == 'Trace - G':
value = self.waveforms['gate'][n]
elif quant.name == 'Trace - Readout trig':
value = self.waveforms['readout_trig']
elif quant.name == 'Trace - Readout I':
value = self.waveforms['readout_iq'].real
elif quant.name == 'Trace - Readout Q':
value = self.waveforms['readout_iq'].imag
# return data as dict with sampling information
dt = 1 / self.sequence_to_waveforms.sample_rate
value = quant.getTraceDict(value, dt=dt)
return value
class Driver(LabberDriver):
"""This class implements a multi-qubit pulse generator."""
def performOpen(self, options={}):
"""Perform the operation of opening the instrument connection."""
# init variables
self.sequence = None
self.sequence_to_waveforms = SequenceToWaveforms()
self.waveforms = {}
# always create a sequence at startup
name = self.getValue('Sequence')
self.sendValueToOther('Sequence', name)
def performSetValue(self, quant, value, sweepRate=0.0, options={}):
"""Perform the Set Value instrument operation."""
# only do something here if changing the sequence type
if quant.name == 'Sequence':
# create new sequence if sequence type changed
new_type = SEQUENCES[value]
if not isinstance(self.sequence, new_type):
# create a new sequence object
if value == 'Custom':
# for custom python files
path = self.getValue('Custom Python file')
(path, modName) = os.path.split(path)
sys.path.append(path)
modName = modName.split('.py')[0] # strip suffix
mod = importlib.import_module(modName)
# the custom sequence class has to be named
# 'CustomSequence'
if not isinstance(self.sequence, mod.CustomSequence):
self.sequence = mod.CustomSequence()
else:
# standard built-in sequence
self.sequence = new_type()
elif (quant.name == 'Custom Python file'
and self.getValue('Sequence') == 'Custom'):
# for custom python files
path = self.getValue('Custom Python file')
(path, modName) = os.path.split(path)
modName = modName.split('.py')[0] # strip suffix
sys.path.append(path)
mod = importlib.import_module(modName)
# the custom sequence class has to be named 'CustomSequence'
if not isinstance(self.sequence, mod.CustomSequence):
self.sequence = mod.CustomSequence()
return value
def performGetValue(self, quant, options={}):
"""Perform the Get Value instrument operation."""
# ignore if no sequence
if self.sequence is None:
return quant.getValue()
# check type of quantity
if (quant.name.startswith('Voltage, QB')
or quant.name.startswith('Single-shot, QB')):
# perform demodulation, check if config is updated
if self.isConfigUpdated():
# update sequence object with current driver configuation
config = self.instrCfg.getValuesDict()
self.sequence.set_parameters(config)
self.sequence_to_waveforms.set_parameters(config)
# get qubit index and waveforms
n = int(quant.name.split(', QB')[1]) - 1
demod_iq = self.getValue('Demodulation - IQ')
if demod_iq:
signal_i = self.getValue('Demodulation - Input I')
signal_q = self.getValue('Demodulation - Input Q')
else:
signal = self.getValue('Demodulation - Input')
ref = self.getValue('Demodulation - Reference')
# perform demodulation
if demod_iq:
value = self.sequence_to_waveforms.readout.demodulate_iq(
n, signal_i, signal_q, ref)
else:
value = self.sequence_to_waveforms.readout.demodulate(
n, signal, ref)
# average values if not single-shot
if not quant.name.startswith('Single-shot, QB'):
value = np.mean(value)
elif quant.isVector():
# traces, check if waveform needs to be re-calculated
if self.isConfigUpdated():
# update sequence object with current driver configuation
config = self.instrCfg.getValuesDict()
self.sequence.set_parameters(config)
self.sequence_to_waveforms.set_parameters(config)
# calcluate waveforms
self.waveforms = self.sequence_to_waveforms.get_waveforms(
self.sequence.get_sequence(config))
# get correct data from waveforms stored in memory
value = self.getWaveformFromMemory(quant)
else:
# for all other cases, do nothing
value = quant.getValue()
return value
def getWaveformFromMemory(self, quant):
"""Return data from already calculated waveforms."""
# check which data to return
if quant.name[-1] in ('1', '2', '3', '4', '5', '6', '7', '8', '9'):
# get name and number of qubit waveform asked for
name = quant.name[:-1]
n = int(quant.name[-1]) - 1
# get correct vector
if name == 'Trace - I':
if self.getValue('Swap IQ'):
value = self.waveforms['xy'][n].imag
else:
value = self.waveforms['xy'][n].real
elif name == 'Trace - Q':
if self.getValue('Swap IQ'):
value = self.waveforms['xy'][n].real
else:
value = self.waveforms['xy'][n].imag
elif name == 'Trace - Z':
value = self.waveforms['z'][n]
elif name == 'Trace - G':
value = self.waveforms['gate'][n]
elif quant.name == 'Trace - Readout trig':
value = self.waveforms['readout_trig']
elif quant.name == 'Trace - Readout I':
value = self.waveforms['readout_iq'].real
elif quant.name == 'Trace - Readout Q':
value = self.waveforms['readout_iq'].imag
# return data as dict with sampling information
dt = 1 / self.sequence_to_waveforms.sample_rate
value = quant.getTraceDict(value, dt=dt)
return value