This subpackage contains drivers to set the OPX as a QCoDeS instrument and synchronize external parameter sweeps with sequences running on the OPX using the QCoDeS do"nd" methods. Since the OPX runs on a unique pulse processor, it can perform complex sequences that QCoDeS do not natively support. For example, acquire data with different formats and sweep parameters on his own. To more fully support the OPX capabilities, this driver is more complex compared to other QCoDeS instruments' drivers.
This package provides tested and ready-to-use drivers and you will find in the examples/Qcodes_drivers section a guide on how to customize the present drivers to better suit your experiments and your way of working with the OPX.
This is where the OPX is defined as a QCoDeS instrument. This file contains several other classes to describe the results acquired with the OPX as QCoDeS parameters as well as setpoint parameters.
Python class inheriting from the QCoDeS Instrument and defining the OPX main driver.
It contains the methods to enable the connection with the OPX by opening a quantum machine manager, update the configuration,
open a quantum machine, simulate, compile and execute QUA programs.
connect_to_qmm(host, port): Enable the connection with the OPX by creating the QuantumMachineManager.get_din(): Parse a standard VISA *IDN? response into an ID dict.connect_message(): Print a standard message on initial connection to an instrument.set_config(config): Update the configuration used by the OPX.open_qm(): Open a quantum machine with a given configuration ready to execute a program.update_qm(): Close and re-open a new quantum machine so that it reloads the configuration in case it has been updated.update_readout_length(readout_element, readout_operation, new_length): Update locally the readout length of a given readout operation and readout element.get_prog(): Get the implemented QUA program.get_res(): Get the results from the OPX.set_sweep_parameters(scanned_axis, setpoints, unit=None, label=None): Set the setpoints based on the sweep parameters.get_measurement_parameter(scale_factor=[(),]): Find the correct Parameter shape based on the stream-processing and return the measurement Parameter. If 'I' and 'Q' are saved in the stream_processing, then the amplitude 'R' and phase 'Phi' will also be returned. Additionally, the default unit for the raw adc traces, and the results from the integration and demodulation methods are automatically converted into Volts. It is however possible to convert it to another unit by specifying a scale factor as an input parameter of the form [(name of the variable, conversion factor, new unit), ], as inscale_factor=[("I", 1235, "pA"), ("Q", 1235, "pA")].execute_prog(prog): Execute a given QUA program and creates a result handle to fetch the results.compile_prog(prog): Compile a given QUA program and stores it under the prog_id attribute.execute_compiled_prog(): Add a compiled program to the current queue and create a result handle to fetch the results.simulate_prog(prog): Simulate a given QUA program and store the simulated waveform into the simulated_wf attribute.plot_simulated_wf(): Plot the simulated waveforms in a new figure using matplotlib.live_plotting(results_to_plot=[]): Enable live fetching and plotting of the data while the QUA program is running. This works only if the averaging is done on the most outer loop and using the.average()method in the stream-processing. See the dedicated section below for a usage example.halt(): Interrupt the current job and halt the running program.close(): Close the quantum machine and tear down the OPX instrument.
Subclass of the QCoDeS MultiParameter class used to define the shapes, setpoints and get_raw() attributes of
specific measurement modes of the OPX as defined in QUA program and Stream Processing.
Classes inheriting from the QCoDeS Parameter class and used to defined the setpoints of the parameter scans performed with QCoDeS.
GeneratedSetPoints: A parameter that generates a setpoint array parametrized with start, stop and number of points.GeneratedSetPointsSpan: A parameter that generates a setpoint array parametrized span, center and number of points.
When integrating the OPX in your QCoDeS framework, there are a few critical points to pay attention to:
- In the QUA program, always insert the pulse sequence under an infinite loop starting with a
pause()command. This is needed to synchronize the OPX sequence with the QCoDeS parameters sweep. - In the QUA program, in the stream processing, make sure to save the results with the
save_all()command. This is needed to synchronize the OPX sequence with the QCoDeS parameters sweep. - In the do'n'd method, make sure to always call
run_expat the beginning to send the QUA program to the OPX,resumeat each iteration to start the OPX pulse sequence,get_measurement_parameter()to define the QCoDeS measurement parameters for this experiment andhaltat the end to close the job. - Make sure to define the setpoint parameters for the OPX scanned axes using
set_sweep_parameters(). Note that "axis1" is the most inner loop and "axis2" is the second loop.
The code block below shows the basic structure for running an experiment with the OPX integrated in the QCoDeS framework.
from qualang_tools.external_frameworks.qcodes.opx_driver import OPX
import os
import qcodes as qc
from qcodes import initialise_or_create_database_at, load_or_create_experiment
from qcodes.utils.dataset.doNd import do2d, do1d, do0d
from qcodes import Parameter
from qm.qua import *
from configuration import *
#####################################
# QCoDeS set-up #
#####################################
db_name = "QM_demo.db" # Database name
sample_name = "demo" # Sample name
exp_name = "OPX_qcodes_drivers" # Experiment name
# Initialize QCoDeS database
db_file_path = os.path.join(os.getcwd(), db_name)
qc.config.core.db_location = db_file_path
initialise_or_create_database_at(db_file_path)
# Initialize QCoDeS experiment
experiment = load_or_create_experiment(
experiment_name=exp_name, sample_name=sample_name
)
# Initialize the QCoDeS station to which instruments will be added
station = qc.Station()
# Create the OPX instrument class
opx_instrument = OPX(config, name="OPX_demo", host="127.0.0.1")
# Add the OPX instrument to the QCoDeS station
station.add_component(opx_instrument)
#####################################
# 2D SWEEP & do0d #
#####################################
n_avg = 100
gate_1_step = 0.01
gate_1_biases = np.arange(-0.2, 0.2, gate_1_step)
gate_1_prefactor = gate_1_step / gate_1_amp
gate_2_step = 0.01
gate_2_biases = np.arange(0, 0.25, gate_2_step)
gate_2_prefactor = gate_2_step / gate_2_amp
def OPX_2d_scan(simulate=False):
with program() as prog:
i = declare(int)
j = declare(int)
n = declare(int)
I = declare(fixed)
Q = declare(fixed)
I_st = declare_stream()
Q_st = declare_stream()
with infinite_loop_():
if not simulate:
pause()
with for_(n, 0, n < n_avg, n + 1):
with for_(i, 0, i < len(gate_1_biases), i + 1):
with if_(i == 0):
play("bias" * amp(0), "gate_1")
with else_():
play("bias" * amp(gate_1_prefactor), "gate_1")
with for_(j, 0, j < len(gate_2_biases), j + 1):
with if_(j == 0):
play("bias" * amp(0), "gate_2")
with else_():
play("bias" * amp(gate_2_prefactor), "gate_2")
wait(200 // 4, "readout_element")
measure("readout", "readout_element", None,
integration.full("cos", I, "out1"),
integration.full("cos", Q, "out2"))
save(I, I_st)
save(Q, Q_st)
ramp_to_zero("gate_2")
ramp_to_zero("gate_1")
with stream_processing():
# Here the averaging is done with .buffer(n_avg).map(FUNCTIONS.average()) so that only the data from the
# same iteration of the qcodes loop are averaged together.
I_st.buffer(len(gate_2_biases)).buffer(len(gate_1_biases)).buffer(n_avg).map(FUNCTIONS.average()).save_all("I")
Q_st.buffer(len(gate_2_biases)).buffer(len(gate_1_biases)).buffer(n_avg).map(FUNCTIONS.average()).save_all("Q")
return prog
# Update the readout length of a given readout operation and readout element.
readout_len = 2_000 # --> 2µs
opx_instrument.update_readout_length("readout_element", "readout", readout_len)
# Specify the readout length used in this experiment to convert the results into Volts and define the setpoints
opx_instrument.readout_pulse_length(readout_len)
# Axis1 is the most inner loop
opx_instrument.set_sweep_parameters("axis1", gate_2_biases, "V", "Gate 2 biases")
# Axis2 is the second loop
opx_instrument.set_sweep_parameters("axis2", gate_1_biases, "V", "Gate 1 biases")
# Add the custom sequence to the OPX
opx_instrument.qua_program = OPX_2d_scan(simulate=True)
# Simulate program
opx_instrument.sim_time(100_000)
opx_instrument.simulate()
opx_instrument.plot_simulated_wf()
# Execute program
opx_instrument.qua_program = OPX_2d_scan(simulate=False)
do0d(
opx_instrument.run_exp,
opx_instrument.resume,
opx_instrument.get_full_data(),
do_plot=True,
exp=experiment
)from qualang_tools.external_frameworks.qcodes.opx_driver import OPX
import os
import qcodes as qc
from qcodes import initialise_or_create_database_at, load_or_create_experiment
from qcodes.utils.dataset.doNd import do2d, do1d, do0d
from qcodes import Parameter
from qm.qua import *
from configuration import *
#####################################
# QCoDeS set-up #
#####################################
db_name = "QM_demo.db" # Database name
sample_name = "demo" # Sample name
exp_name = "OPX_qcodes_drivers" # Experiment name
# Initialize QCoDeS database
db_file_path = os.path.join(os.getcwd(), db_name)
qc.config.core.db_location = db_file_path
initialise_or_create_database_at(db_file_path)
# Initialize QCoDeS experiment
experiment = load_or_create_experiment(
experiment_name=exp_name, sample_name=sample_name
)
# Initialize the QCoDeS station to which instruments will be added
station = qc.Station()
# Create the OPX instrument class
opx_instrument = OPX(config, name="OPX_demo", host="127.0.0.1")
# Add the OPX instrument to the QCoDeS station
station.add_component(opx_instrument)
#####################################
# 2D SWEEP & do0d #
#####################################
n_avg = 1000
gate_1_step = 0.01
gate_1_biases = np.arange(-0.2, 0.2, gate_1_step)
gate_1_prefactor = gate_1_step / gate_1_amp
gate_2_step = 0.01
gate_2_biases = np.arange(0, 0.25, gate_2_step)
gate_2_prefactor = gate_2_step / gate_2_amp
def OPX_2d_scan_liveplot(simulate=False):
with program() as prog:
i = declare(int)
j = declare(int)
n = declare(int)
I = declare(fixed)
Q = declare(fixed)
I_st = declare_stream()
Q_st = declare_stream()
# The infinite loop was removed, otherwise the live plotting never stops
# and the python console remains busy forever.
if not simulate:
pause()
with for_(n, 0, n < n_avg, n+1):
with for_(i, 0, i < len(gate_1_biases), i + 1):
with if_(i == 0):
play("bias" * amp(0), "gate_1")
with else_():
play("bias" * amp(gate_1_prefactor), "gate_1")
with for_(j, 0, j < len(gate_2_biases), j + 1):
with if_(j == 0):
play("bias" * amp(0), "gate_2")
with else_():
play("bias" * amp(gate_2_prefactor), "gate_2")
wait(200 // 4, "readout_element")
measure("readout", "readout_element", None,
integration.full("cos", I, "out1"),
integration.full("cos", Q, "out2"))
save(I, I_st)
save(Q, Q_st)
ramp_to_zero("gate_2")
ramp_to_zero("gate_1")
with stream_processing():
# Note that `.buffer(n_avg).map(FUNCTIONS.average())` was replaced by `.average`,
# since the live plotting feature will only work when used outside of dond as shown below.
I_st.buffer(len(gate_2_biases)).buffer(len(gate_1_biases)).average().save_all("I")
Q_st.buffer(len(gate_2_biases)).buffer(len(gate_1_biases)).average().save_all("Q")
return prog
# Update the readout length of a given readout operation and readout element.
readout_len = 2_000 # --> 2µs
opx_instrument.update_readout_length("readout_element", "readout", readout_len)
# Specify the readout length used in this experiment to convert the results into Volts and define the setpoints
opx_instrument.readout_pulse_length(readout_len)
# Axis1 is the most inner loop
opx_instrument.set_sweep_parameters("axis1", gate_2_biases, "V", "Gate 2 biases")
# Axis2 is the second loop
opx_instrument.set_sweep_parameters("axis2", gate_1_biases, "V", "Gate 1 biases")
# Add the custom sequence to the OPX
opx_instrument.qua_program = OPX_2d_scan_liveplot(simulate=True)
# Simulate program
opx_instrument.sim_time(100_000)
opx_instrument.simulate()
opx_instrument.plot_simulated_wf()
# Execute program
opx_instrument.qua_program = OPX_2d_scan_liveplot(simulate=False)
# Compile the QUA program and execute it
opx_instrument.run_exp()
# Exit the pause() statement and start the sequence
opx_instrument.resume()
# Fetch the data in real-time and plot them
opx_instrument.live_plotting(["I", "Q"])
# Update the data counter to save the last dataset to the qcodes database
opx_instrument.counter = n_avg
# Store the results in the qcodes database
do0d(
opx_instrument.get_measurement_parameter(),
do_plot=True,
exp=experiment,
)More details and advices to create your own experiment driver can be found in the examples section: https://github.com/qua-platform/py-qua-tools/tree/main/examples/Qcodes_drivers