def getter(*names_units: Tuple) -> Callable:
"""
Args:
names_units
List of tuples with parameter names and units (and optionally
'paramtype' that defines how the data is saved),
e.g. [("gate", "V"), ("Isd", "A", "array")]
Returns:
A decorator. The decorated function returns a callable and a parameter
table. The callable calls the decorated function which should return
measurement values.
For more information about 'paramtype' argument, see `register_parameter`
method of `Measurement` class in QCoDeS.
"""
table = param_table.add(_generate_tables(names_units))
def decorator(func: Callable) -> MeasureFunction:
def inner() -> dict:
results = np.atleast_1d(func())
return {k[0]: v for k, v in zip(names_units, results)}
return MeasureFunction(inner, table.copy())
return decorator
def getter(*names_units: Tuple) -> Callable:
"""
Args:
names_units: List of tuples with parameter names and units, e.g.
[("gate", "V"), ("Isd", "A")]
Returns:
A decorator. The decorated function returns a callable and a parameter
table. The callable calls the decorated function which should return
measurement values.
"""
table = param_table.add(_generate_tables(names_units))
def decorator(func: Callable) -> Callable:
def inner() -> Tuple[Callable, ParamTable]:
def wrapper() -> dict:
results = np.atleast_1d(func())
return {k[0]: v for k, v in zip(names_units, results)}
return wrapper, table
inner.getter_setter_decorated = True
return inner
return decorator
def __init__(self, *sweep_objects: BaseSweepObject) -> None:
super().__init__()
self._sweep_objects = sweep_objects
self._parameter_table = param_table.add(
[so.parameter_table for so in sweep_objects])
self._measurable = any([so.measurable for so in sweep_objects])
def hardsweep(ind: List[Tuple], dep: List[Tuple]) -> Callable:
"""
Args:
ind: List of independent parameters, defined as tuples of names and
units.
dep: List of dependent parameters, defined as tuples of names and
units.
Returns:
A decorator which returns a sweep object, which can be directly used
in pytopo.do_experiment to run an experiment.
Example:
>>> import pytopo
>>> @hardsweep(ind=[("x", "V"), ("y", "V")], dep=[("i", "A")])
>>> def some_function():
>>> ... # Some code setting up instruments and performing measurements
>>> ... return set_points, measurements
>>> pytopo.do_experiement("name/sample", some_function)
Since we have defined to independent parameters, `x` and `y`, the
returned `set_points` should be a 2-by-N numpy array, where `N` is the
number of set points. We have one dependent parameters, `i`, therefor
`measurements` is an 1-by-N numpy array.
The signature of `some_function` can be anything the user desires.
However, the return value of this function has to consist of two numpy
arrays with the aforementioned shape.
Please see pytopo/sweep/docs/hardsweep.ipynb for a more elaborate
example
"""
# If we have two independent parameters, say `x` and `y`, then we are
# sampling in an inner product space spanned by two axes. Hence we need to
# use the `prod` operator to generate the appropriate table
ind_table = param_table.prod(_generate_tables(ind))
# Each dependent parameter adds a seperate measurement. Hence we need to
# use the `add` operator
dep_table = param_table.add(_generate_tables(dep))
# Each dependent parameter needs to be nested in the table of independent
# parameters. This is performed with the `prod` operator.
table = param_table.prod([ind_table, dep_table])
def decorator(func: Callable) -> Callable:
def inner(*args, **kwargs) -> IteratorSweep:
def wrapper() -> dict:
spoints, measurements = func(*args, **kwargs)
spoints = np.atleast_2d(spoints)
measurements = np.atleast_2d(measurements)
if spoints.shape[0] != len(ind) or \
measurements.shape[0] != len(dep):
raise ValueError("The number of points or measurements "
"returned does not match the number of "
"dependent and/or independent parameters")
for spoint, measurement in zip(spoints.T, measurements.T):
res = {k[0]: v for k, v in zip(ind, spoint)}
res.update({k[0]: v for k, v in zip(dep, measurement)})
yield res
sweep_object = IteratorSweep(wrapper,
parameter_table=table.copy(),
measurable=True)
return sweep_object
return inner
return decorator