Пример #1
0
    def __init__(self, name: str = None,
                 use_as_attributes: bool = False,
                 log_changes: bool = True,
                 simplify_snapshot: bool = False,
                 parent: Union['ParameterNode', bool] = None,
                 **kwargs):
        # Move deepcopy method to the instance scope, since it will temporarily
        # delete its own method during copying (see ParameterNode.__deepcopy__)
        self.__deepcopy__ = partial(__deepcopy__, self)

        self.use_as_attributes = use_as_attributes
        self.log_changes = log_changes
        self.simplify_snapshot = simplify_snapshot

        if name is not None:
            self.name = name

        # Parent ParameterNode. Can also be None or False.
        # If set to None, it will be set the next time this node is attached as
        # an attribute of another Node.
        # If False, setting this Node to an attribute of another Node will not
        # set this attribute
        self.parent = parent

        self.parameters = DotDict()
        self.parameter_nodes = DotDict()
        self.functions = {}
        self.submodules = {}

        super().__init__(**kwargs)

        self._meta_attrs = ['name']
Пример #2
0
    def __init__(self,
                 name: str,
                 folder: str = None,
                 parent: SubConfig = None,
                 config: dict = None,
                 save_as_dir: bool = None):
        self.initializing = True
        self._mirrored_config_attrs = {}
        self._inherited_configs = []

        SubConfig.__init__(self,
                           name=name,
                           folder=folder,
                           parent=parent,
                           save_as_dir=save_as_dir)
        DotDict.__init__(self)
        SignalEmitter.__init__(self, initialize_signal=False)

        if config is not None:
            update_dict(self, config)
        elif folder is not None:
            self.load()

        if self.parent is None:
            self._attach_mirrored_items()
Пример #3
0
class PerformanceTimer():
    max_records = 100

    def __init__(self):
        self.timings = DotDict()

    def __getitem__(self, key):
        val = self.timings.__getitem__(key)
        return self._timing_to_str(val)

    def __repr__(self):
        return pprint.pformat(self._timings_to_str(self.timings), indent=2)

    def clear(self):
        self.timings.clear()

    def _timing_to_str(self, val):
        mean_val = np.mean(val)
        exponent, prefactor = get_exponent(mean_val)
        factor = np.power(10., exponent)

        return f'{mean_val / factor:.3g}+-{np.abs(np.std(val))/factor:.3g} {prefactor}s'

    def _timings_to_str(self, d: dict):

        timings_str = DotDict()
        for key, val in d.items():
            if isinstance(val, dict):
                timings_str[key] = self._timings_to_str(val)
            else:
                timings_str[key] = self._timing_to_str(val)

        return timings_str

    @contextmanager
    def record(self, key, val=None):
        if isinstance(key, str):
            timing_list = self.timings.setdefault(key, [])
        elif isinstance(key, (list)):
            *parent_keys, subkey = key
            d = self.timings.create_dicts(*parent_keys)
            timing_list = d.setdefault(subkey, [])
        else:
            raise ValueError('Key must be str or list/tuple')

        if val is not None:
            timing_list.append(val)
        else:
            t0 = time.perf_counter()
            yield
            t1 = time.perf_counter()
            timing_list.append(t1 - t0)

        # Optionally remove oldest elements
        for _ in range(len(timing_list) - self.max_records):
            timing_list.pop(0)
Пример #4
0
 def __contains__(self, key):
     if DotDict.__contains__(self, key):
         return True
     elif DotDict.__contains__(self, 'inherit'):
         try:
             if self['inherit'].startswith('config:') or \
                     self['inherit'].startswith('environment:'):
                 return key in self[self['inherit']]
             else:
                 return key in self.parent[self['inherit']]
         except KeyError:
             return False
     else:
         return False
Пример #5
0
    def _add_DC_waveform(
        self,
        channel_name: str,
        t_start: float,
        t_stop: float,
        amplitude: float,
        sample_rate: float,
        pulse_name="DC",
    ) -> List:
        # We fake a DC pulse for improved performance
        DC_pulse = DotDict(
            dict(
                t_start=t_start,
                t_stop=t_stop,
                duration=round(t_stop - t_start, 11),
                amplitude=amplitude,
            ))
        waveform = DCPulseImplementation.implement(pulse=DC_pulse,
                                                   sample_rate=sample_rate)
        sequence_steps = self.add_pulse_waveforms(
            channel_name,
            **waveform,
            t_stop=DC_pulse.t_stop,
            sample_rate=sample_rate,
            pulse_name=pulse_name,
        )

        return sequence_steps
 def __init__(self, name, **kwargs):
     super().__init__(
         name=name,
         names=(),
         snapshot_value=False,
         properties_attrs=["analyses"],
         **kwargs,
     )
     self.results = DotDict()
Пример #7
0
    def __getitem__(self, key):
        if key.startswith('config:'):
            if self.parent is not None:
                # Let parent config deal with this
                return self.parent[key]
            elif key == 'config:':
                return self
            else:
                return self[key.replace('config:', '')]
        elif key.startswith('environment:'):
            if self.parent is None:
                if silq.environment is None:
                    environment_config = self
                else:
                    environment_config = self[silq.environment]

                if key == 'environment:':
                    return environment_config
                else:
                    return environment_config[key.replace('environment:', '')]
            else:
                # Pass environment:path along to parent
                return self.parent[key]
        elif DotDict.__contains__(self, key):
            val = DotDict.__getitem__(self, key)
            if key == 'inherit':
                return val
            elif isinstance(val, str) and \
                    (val.startswith('config:') or val.startswith('environment:')):
                try:
                    return self[val]
                except KeyError:
                    raise KeyError(
                        f"Couldn't retrieve mirrored key {key} -> {val}")
            else:
                return val
        elif 'inherit' in self:
            if self['inherit'].startswith('config:') or \
                    self['inherit'].startswith('environment:'):
                return self[self['inherit']][key]
            else:
                return self.parent[self['inherit']][key]
        else:
            raise KeyError(f"Couldn't retrieve key {key}")
Пример #8
0
    def _timings_to_str(self, d: dict):

        timings_str = DotDict()
        for key, val in d.items():
            if isinstance(val, dict):
                timings_str[key] = self._timings_to_str(val)
            else:
                timings_str[key] = self._timing_to_str(val)

        return timings_str
    def analyse(self, traces=None, plot=False, plot_high_low=False):
        """Analyse ESR traces.

        If there is only one ESR pulse, returns ``up_proportion_{pulse.name}``.
        If there are several ESR pulses, adds a zero-based suffix at the end for
        each ESR pulse. If ``ESRParameter.EPR['enabled'] == True``, the results
        from `analyse_EPR` are also added, as well as ``contrast_{pulse.name}``
        (plus a suffix if there are several ESR pulses).
        """
        if traces is None:
            traces = self.traces
        # Convert to DotDict so we can access nested pulse sequences
        traces = DotDict(traces)

        results = DotDict()

        # First analyse EPR because we want to use dark_counts
        if "EPR" in traces:
            results["EPR"] = self.analyses.EPR.analyse(
                empty_traces=traces[self.EPR[0].full_name]["output"],
                plunge_traces=traces[self.EPR[1].full_name]["output"],
                read_traces=traces[self.EPR[2].full_name]["output"],
                plot=plot,
            )
            dark_counts = results["EPR"]["dark_counts"]
        else:
            dark_counts = None

        for name, analysis in self.analyses.parameter_nodes.items():
            if name == 'EPR':
                continue
            if isinstance(analysis, AnalyseElectronReadout):
                results[name] = analysis.analyse(
                    traces=traces[name],
                    dark_counts=dark_counts,
                    plot=plot,
                    plot_high_low=plot_high_low
                )
            else:
                raise SyntaxError(f'Cannot process analysis {name} {type(analysis)}')

        self.results = results
        return results
Пример #10
0
    def test_mask_config(self):
        c = DotDict({"x": 1})

        with Measurement("mask_parameter") as msmt:
            self.assertEqual(c.x, 1)
            msmt.mask(c, x=2)
            self.assertEqual(len(msmt._masked_properties), 1)
            self.assertDictEqual(
                msmt._masked_properties[0],
                {"type": "key", "obj": c, "key": "x", "original_value": 1, "value": 2},
            )

            for k in Sweep(range(10), "repetition"):
                msmt.measure({"a": 3, "b": 4}, "acquire_values")
                self.assertEqual(c.x, 2)
            self.assertEqual(c.x, 2)

        self.assertEqual(c.x, 1)
        self.assertEqual(len(msmt._masked_properties), 0)
Пример #11
0
    def test_double_mask(self):
        d = DotDict({"x": 1})

        with Measurement("mask_parameter") as msmt:
            self.assertEqual(d.x, 1)
            msmt.mask(d, x=2)
            self.assertEqual(d.x, 2)
            msmt.mask(d, x=3)
            self.assertEqual(d.x, 3)
            self.assertEqual(len(msmt._masked_properties), 2)
            self.assertListEqual(
                msmt._masked_properties,
                [
                    {
                        "type": "key",
                        "obj": d,
                        "key": "x",
                        "original_value": 1,
                        "value": 2,
                    },
                    {
                        "type": "key",
                        "obj": d,
                        "key": "x",
                        "original_value": 2,
                        "value": 3,
                    },
                ],
            )

            for k in Sweep(range(10), "repetition"):
                msmt.measure({"a": 3, "b": 4}, "acquire_values")
                self.assertEqual(d.x, 3)
            self.assertEqual(d.x, 3)

        self.assertEqual(d.x, 1)
        self.assertEqual(len(msmt._masked_properties), 0)
Пример #12
0
    def __init__(self, update_function=None, **kwargs):
        self.update_function = update_function
        super().__init__()

        for key, val in kwargs.items():
            DotDict.__setitem__(self, key, val)
Пример #13
0
    def test_mask_attr_does_not_exist(self):
        c = DotDict()

        with Measurement("mask_parameter") as msmt:
            with self.assertRaises(KeyError):
                msmt.mask(c, x=2)
Пример #14
0
class ParameterNode(Metadatable, DelegateAttributes, metaclass=ParameterNodeMetaClass):
    """ Container for parameters

    The ParameterNode is a container for `Parameters`, and is primarily used for
    an `Instrument`.

    Args:
        name: Optional name for parameter node
        use_as_attributes: Treat parameters as attributes (see below)
        log_changes: Log all changes of parameter values as debug messages
        simplify_snapshot: Snapshot contains simplified parameter snapshots

    A parameter can be added to a ParameterNode by settings its attribute:
    ``parameter_node.new_parameter = Parameter()``
    The name of the parameter is set to the attribute name

    Once a parameter has been added, its value can be set/get depending on the
    arg use_as_attributes. If use_as_attributes is False, calling
    ``parameter_node.new_parameter`` returns the Parameter object.
    The parameter is set using ``parameter_node.new_parameter(value)`` and
    retrieved via ``parameter_node.new_parameter()`` (same as you would for a
    parameter that does not belong to a Node).

    If ``use_as_attributes`` is True, its value can be set as such:
    ``parameter_node.new_parameter = 42``
    Note that this doesn't replace the parameter by 42, but instead sets the
    value of the parameter.

     Similarly, its value can be returned by accessing the attribute
     ``parameter_node.new_parameter`` (returns 42)
     Again, this doesn't return the parameter, but its value

    The parameter object can then be accessed via ``parameter_node['new_parameter']``
    """

    parameters = {}
    parameter_nodes = {}    #
    # attributes to delegate from dict attributes, for example:
    # instrument.someparam === instrument.parameters['someparam']
    delegate_attr_dicts = ['parameters', 'parameter_nodes', 'functions',
                           'submodules']

    _deepcopy_skip_parameters = []

    def __init__(self, name: str = None,
                 use_as_attributes: bool = False,
                 log_changes: bool = True,
                 simplify_snapshot: bool = False,
                 parent: Union['ParameterNode', bool] = None,
                 **kwargs):
        # Move deepcopy method to the instance scope, since it will temporarily
        # delete its own method during copying (see ParameterNode.__deepcopy__)
        self.__deepcopy__ = partial(__deepcopy__, self)

        self.use_as_attributes = use_as_attributes
        self.log_changes = log_changes
        self.simplify_snapshot = simplify_snapshot

        if name is not None:
            self.name = name

        # Parent ParameterNode. Can also be None or False.
        # If set to None, it will be set the next time this node is attached as
        # an attribute of another Node.
        # If False, setting this Node to an attribute of another Node will not
        # set this attribute
        self.parent = parent

        self.parameters = DotDict()
        self.parameter_nodes = DotDict()
        self.functions = {}
        self.submodules = {}

        super().__init__(**kwargs)

        self._meta_attrs = ['name']

    def __str__(self):
        s = ''
        if getattr(self, 'name', None):
            if self.parent and getattr(self.parent, 'name', None) not in [None, '']:
                s += f'{self.parent.name}_'
            if isinstance(self.name, _BaseParameter):
                s += f'{self.name()}'
            else:
                s += f'{self.name}'
        return s

    def __repr__(self):
        repr_str = f'ParameterNode {str(self) + " " if str(self) else ""}containing '
        if self.parameter_nodes:
            repr_str += f'{len(self.parameter_nodes)} node{"s" if len(self.parameter_nodes) != 1 else ""}:\n'
            for parameter_node in self.parameter_nodes:
                repr_str += f'\t{parameter_node}\n'
        if self.parameters:
            repr_str += f'{len(self.parameters)} parameter{"s" if len(self.parameters) != 1 else ""}:\n'
            for parameter in self.parameters:
                    repr_str += f'\t{parameter}\n'
        return repr_str

    def __getattr__(self, attr):
        if attr == 'use_as_attributes':
            return super().__getattr__(attr)
        elif attr in self.parameters:
            parameter = self.parameters[attr]
            if self.use_as_attributes:
                # Perform get and return value
                return parameter()
            else:
                # Return parameter instance
                return parameter
        elif attr in self.parameter_nodes:
            return self.parameter_nodes[attr]
        else:
            return super().__getattr__(attr)

    def __setattr__(self, attr, val):
        if attr == 'parent':
            super().__setattr__(attr, val)
        elif isinstance(val, _BaseParameter):
            # Delete attr if it already exists so we can delegate param attribute
            if attr in self.__dict__:
                delattr(self, attr)

            self.parameters[attr] = val
            if val.parent is None:  # Attach self as parent if not already set
                val.parent = self

            if hasattr(self, 'multiple_senders'):
                val.multiple_senders = self.multiple_senders

            if attr in self._parameter_decorators:
                # Some methods have been defined in the ParameterNode as methods
                # Using the @parameter decorator.
                self._attach_parameter_decorators(
                    parameter=val,
                    decorator_methods=self._parameter_decorators[attr])
            if val.name == 'None':
                # Parameter has been created without name, update name to attr
                val.name = attr
                if val.label is None:
                    # For label, convert underscores to spaces and capitalize
                    label = attr.replace('_', ' ')
                    val.label = label[0].capitalize() + label[1:]
            val.log_changes = self.log_changes
        elif isinstance(val, ParameterNode):
            # Check if attribute is an @property
            if isinstance(getattr(type(self), attr, None), property):
                super().__setattr__(attr, val)
                return

            # Delete attr if it already exists so we can delegate node attribute
            if attr in self.__dict__:
                delattr(self, attr)

            self.parameter_nodes[attr] = val
            if val.parent is None:  # Attach self as parent if not already set
                val.parent = self
            if not hasattr(val, 'name'):
                # Update nested ParameterNode name
                val.name = attr
            val.log_changes = self.log_changes
        elif attr in self.parameters:
            # Set parameter value
            self.parameters[attr](val)
        else:
            super().__setattr__(attr, val)

    def __copy__(self):
        """Copy method for ParameterNode, invoked by copy.copy(parameter_node).
        """
        rv = self.__reduce_ex__(4)
        self_copy = _reconstruct(self, None, *rv)

        self_copy.__deepcopy__ = partial(__deepcopy__, self_copy)
        self_copy.parent = None  # By default no parent for copied nodes

        self_copy.parameters = {}
        for parameter_name, parameter in self.parameters.items():
            parameter_copy = copy(parameter)
            self_copy.parameters[parameter_name] = parameter_copy
            parameter_copy.parent = self_copy

        # Attach parameter decorators, done in a second loop because the
        # decorators may call parameters, and so all parameters must exist
        for parameter_name, parameter_decorators in self._parameter_decorators.items():
            parameter = self_copy.parameters[parameter_name]
            self_copy._attach_parameter_decorators(parameter, parameter_decorators)

        self_copy.parameter_nodes = {}
        for node_name, parameter_node in self_copy.parameter_nodes.items():
            parameter_node_copy = copy(parameter_node)
            self_copy.parameter_nodes[node_name] = parameter_node_copy
            parameter_node_copy.parent = self_copy

        return self_copy

    def __getitem__(self, key):
        """Delegate instrument['name'] to parameter or function 'name'."""
        try:
            return self.parameters[key]
        except KeyError:
            pass
        try:
            return self.functions[key]
        except KeyError:
            pass
        return super().__getitem__(key)

    def __dir__(self):
        # Add parameters to dir
        items = super().__dir__()
        items.extend(self.parameters)
        items.extend(self.parameter_nodes)
        return items

    def __eq__(self, other):
        return self.matches_parameter_node(other)

    def __getstate__(self):
        d = copy(self.__dict__)
        d['signal'] = Signal()
        d['parent'] = None
        return d

    def _attach_parameter_decorators(self,
                                     parameter: _BaseParameter,
                                     decorator_methods: Dict[str, Callable]):
        """Attaches @parameter decorators to a parameter

        Args:
            parameter: Parameter to attach decorators to
            decorator_methods: Decorator methods to attach to parameter
            """
        for param_attr, param_method in decorator_methods.items():
            method_with_args = partial(param_method, self, parameter)
            if param_attr == 'get':
                parameter.get_raw = method_with_args
                if parameter.wrap_get:
                    parameter.get = parameter._wrap_get(parameter.get_raw)
                else:
                    parameter.get = parameter.get_raw
            elif param_attr == 'set':
                parameter.set_raw = method_with_args
                if parameter.wrap_set:
                    parameter.set = parameter._wrap_set(parameter.set_raw)
                else:
                    parameter.set = parameter.set_raw
            else:
                setattr(parameter, param_attr, method_with_args)
        # perform a set without evaluating, which saves the value,
        # ensuring that new modifications such as the set_parser are
        # taken into account
        if hasattr(parameter, 'set') and parameter.raw_value is not None and parameter.wrap_set:
            parameter.set(copy(parameter.get_latest()), evaluate=False)

    def add_function(self, name: str, **kwargs):
        """ Bind one Function to this parameter node.

        Instrument subclasses can call this repeatedly in their ``__init__``
        for every real function of the instrument.

        This functionality is meant for simple cases, principally things that
        map to simple commands like '\*RST' (reset) or those with just a few
        arguments. It requires a fixed argument count, and positional args
        only. If your case is more complicated, you're probably better off
        simply making a new method in your ``Instrument`` subclass definition.

        Args:
            name: how the Function will be stored within
            ``instrument.Functions`` and also how you  address it using the
            shortcut methods: ``parameter_node.call(func_name, *args)`` etc.

            **kwargs: constructor kwargs for ``Function``

        Raises:
            KeyError: if this instrument already has a function with this
                name.
        """
        if name in self.functions:
            raise KeyError('Duplicate function name {}'.format(name))
        func = Function(name=name, instrument=self, **kwargs)
        self.functions[name] = func

    def add_submodule(self, name: str, submodule: Metadatable):
        """ Bind one submodule to this instrument.

        Instrument subclasses can call this repeatedly in their ``__init__``
        method for every submodule of the instrument.

        Submodules can effectively be considered as instruments within the main
        instrument, and should at minimum be snapshottable. For example, they
        can be used to either store logical groupings of parameters, which may
        or may not be repeated, or channel lists.

        Args:
            name: how the submodule will be stored within
                `instrument.submodules` and also how it can be addressed.

            submodule: The submodule to be stored.

        Raises:
            KeyError: if this instrument already contains a submodule with this
                name.
            TypeError: if the submodule that we are trying to add is not an
                instance of an Metadatable object.
        """
        if name in self.submodules:
            raise KeyError('Duplicate submodule name {}'.format(name))
        if not isinstance(submodule, Metadatable):
            raise TypeError('Submodules must be metadatable.')
        self.submodules[name] = submodule

    def snapshot_base(self, update: bool=False,
                      params_to_skip_update: Sequence[str]=None,
                      skip_parameters: Sequence[str] = (),
                      skip_parameter_nodes: Sequence[str] = ()):
        """
        State of the instrument as a JSON-compatible dict.

        Args:
            update (bool): If True, update the state by querying the
                instrument. If False, just use the latest values in memory.
            params_to_skip_update: List of parameter names that will be skipped
                in update even if update is True. This is useful if you have
                parameters that are slow to update but can be updated in a
                different way (as in the qdac)
            skip_parameters: Names of parameters to skip from snapshot
            skip_parameter_nodes: Names of parameter nodes to skip from snapshot

        Returns:
            dict: base snapshot
        """
        if self.simplify_snapshot:
            snap = {"__class__": full_class(self)}
            if self.functions:
                snap["functions"] = {name: func.snapshot(update=update)
                                     for name, func in self.functions.items()}
            if self.submodules:
                snap["submodules"] = {name: subm.snapshot(update=update)
                                      for name, subm in self.submodules.items()}
            for parameter_name, parameter in self.parameters.items():
                if parameter_name in skip_parameters:
                    continue
                parameter_snapshot = parameter.snapshot()
                if 'unit' in parameter_snapshot:
                    parameter_name = f'{parameter_name} ({parameter_snapshot["unit"]})'
                if parameter._snapshot_value:
                    snap[parameter_name] = parameter_snapshot['value']
                else:
                    snap[parameter_name] = parameter_snapshot
            for parameter_node_name, parameter_node in self.parameter_nodes.items():
                if parameter_node_name in skip_parameter_nodes:
                    continue
                snap[parameter_node_name] = parameter_node.snapshot()
        else:
            snap = {
                "functions": {name: func.snapshot(update=update)
                              for name, func in self.functions.items()},
                "submodules": {name: subm.snapshot(update=update)
                               for name, subm in self.submodules.items()},
                "__class__": full_class(self),
                "parameters": {},
                "parameter_nodes": {
                    name: node.snapshot()
                    for name, node in self.parameter_nodes.items()
                    if name not in skip_parameter_nodes
                }
            }

            for name, param in self.parameters.items():
                if name in skip_parameters:
                    continue
                update = update
                if params_to_skip_update and name in params_to_skip_update:
                    update = False
                try:
                    snap['parameters'][name] = param.snapshot(
                        update=update, simplify=self.simplify_snapshot)
                except:
                    logger.debug(f"Snapshot: Could not update parameter: {name}")
                    snap['parameters'][name] = param.snapshot(
                        update=False, simplify=self.simplify_snapshot)
            for attr in set(self._meta_attrs):
                if hasattr(self, attr):
                    snap[attr] = getattr(self, attr)

        return snap

    def to_dict(self, get_latest: bool = True) -> Dict:
        """Generate a dictionary representation of the parameter node

        This method is similar to `ParameterNode.snapshot()`, but more compact.
        Basically, every key is a parameter/parameternode name.
        For a Parameter, the dict value is the parameter's value
        For a ParameterNode, the value is another nested dict.

        Args:
            get_latest: Whether to add the latest value of a parameter,
                or perform a parameter.get() to get the value.

        Returns:
            Dictionary representation of the parameter node
        """
        if get_latest:
            parameters_dict = {
                name: parameter.get_latest() for name, parameter in self.parameters.items()
            }
        else:
            parameters_dict = {
                name: parameter() for name, parameter in self.parameters.items()
            }
        for key, val in parameters_dict.items():
            if isinstance(val, (list, tuple)):
                parameters_dict[key] = [
                    elem if not isinstance(elem, ParameterNode) else elem.to_dict()
                    for elem in val
                ]

        parameter_nodes_dict = {
            name: parameter_node.to_dict(get_latest=get_latest)
            for name, parameter_node in self.parameter_nodes.items()
        }

        combined_dict = {**parameters_dict, **parameter_nodes_dict}
        return combined_dict

    def matches_parameter_node(self,
                                other: Any,
                                exclude_parameters: List[str] = [],
                                exclude_parameter_nodes: List[str] = []) -> bool:
        """Tests if the parameters of another node matches this one.

        Returns:
            False if any parameter values aren't equal
            False if the compared object is not exactly the same class.
            False if the number of parameters doesn't match
            True if all parameter values match
        """
        if self is other:
            return True

        if other.__class__ != self.__class__:
            return False
        elif len(self.parameters) != len(other.parameters):
            return False
        elif len(self.parameter_nodes) != len(other.parameter_nodes):
            return False

        for parameter_name, parameter in self.parameters.items():
            if parameter_name in exclude_parameters:
                continue
            elif not parameter_name in other.parameters \
                    or parameter.get_latest() != other.parameters[parameter_name].get_latest():
                return False

        for parameter_node_name, parameter_node in self.parameter_nodes.items():
            if parameter_node_name in exclude_parameter_nodes:
                continue
            elif not parameter_node_name in other.parameter_nodes \
                    or not parameter_node == other.parameter_nodes[parameter_node_name]:
                return False
        return True

    def sweep(self, parameter_name: str, start=None, stop=None, step=None,
              num=None, **kwargs):
        """ Sweep a parameter in the parameter node

        The following lines are identical:

        >>> parameter_node.param.sweep(start=0, stop=10, step=1)
        >>> parameter_node['param'].sweep(start=0, stop=10, step=1)
        >>> parameter_node.sweep('param', start=0, stop=10, step=1)

        Using the parameter node's sweep method is the recommended method,
        especially if parameter_node.use_as_attributes == True.

        Args:
            parameter_name: Name of parameter to sweep
            start: Sweep start value. Does not need to be set if window is set
            stop: Sweep stop value. Does not need to be set if window is set
            step: Optional sweep step. Does not need to be set if num or
                step_percentage is set
            num: Optional number of sweep values between start and stop. Does
                not need to be set if step or step_percentage is set
            **kwargs: Additional sweep kwargs, for SilQ QCoDeS these are:
                window: Optional sweep window around current value.
                    If set, start and stop do not need to be set
                step_percentage: Optional step percentage, calculated from silq
                    config (needs work, and does not work in unforked QCoDeS)
        """
        return self.parameters[parameter_name].sweep(start=start, stop=stop,
                                                     step=step, num=num,
                                                     **kwargs)

    def print_snapshot(self,
                       update: bool = False,
                       max_chars: int = 80,
                       level=0):
        """ Prints a readable version of the snapshot.

        The readable snapshot includes the name, value and unit of each
        parameter.
        A convenience function to quickly get an overview of the parameter node.

        Args:
            update: If True, update the state by querying the
                instrument. If False, just use the latest values in memory.
                This argument gets passed to the snapshot function.
            max_chars: the maximum number of characters per line. The
                readable snapshot will be cropped if this value is exceeded.
                Defaults to 80 to be consistent with default terminal width.
        """

        pretabs = '\t'*level  # Number of tabs at the start of each line

        floating_types = (float, np.integer, np.floating)

        try:
            # Temporarily set simplify_snapshot to False for snapshotting
            simplify_snapshot = self.simplify_snapshot
            self.simplify_snapshot = False

            snapshot = self.snapshot(update=update)
        finally:
            self.simplify_snapshot = simplify_snapshot


        if snapshot['parameters']:
            par_lengths = [len(p) for p in snapshot['parameters']]
            # Min of 50 to prevent a long parameter name to break this function
            par_field_len = min(max(par_lengths)+1, 50)
        else:
            par_field_len = 50

        if str(self):
            print(pretabs + f'{self} :')
        print(pretabs + '{0:<{1}}'.format('\tparameter ', par_field_len) + 'value')
        print(pretabs + '-'*max_chars)
        for param_name, param_snapshot in sorted(snapshot['parameters'].items()):
            name = param_snapshot['name']
            msg = '{0:<{1}}:'.format(name, par_field_len)

            # in case of e.g. ArrayParameters, that usually have
            # snapshot_value == False, the parameter may not have
            # a value in the snapshot
            val = param_snapshot.get('value', 'Not available')
            if isinstance(val, floating_types):
                msg += f'\t{val:.5g} '
            else:
                msg += f'\t{val} '

            unit = param_snapshot.get('unit', None)
            if unit is None:
                # this may be a multi parameter
                unit = param_snapshot.get('units', None)
            if unit:
                msg += f'({unit})'

            # Truncate the message if it is longer than max length
            if len(msg) > max_chars and not max_chars == -1:
                msg = msg[0:max_chars-3] + '...'
            print(pretabs + msg)

        for submodule in self.submodules.values():
            if hasattr(submodule, '_channels'):
                if submodule._snapshotable:
                    for channel in submodule._channels:
                        channel.print_snapshot()
            else:
                submodule.print_snapshot(update, max_chars)

        for parameter_node in self.parameter_nodes.values():
            print(f'\n{parameter_node.name}')
            parameter_node.print_snapshot(level=level+1)

    def call(self, func_name: str, *args, **kwargs):
        """ Shortcut for calling a function from its name.

        Args:
            func_name: The name of a function of this instrument.
            *args: any arguments to the function.
            **kwargs: any keyword arguments to the function.

        Returns:
            any: The return value of the function.
        """
        return self.functions[func_name].call(*args, **kwargs)

    def validate_status(self,
                        verbose: bool = False):
        """ Validate the values of all gettable parameters

        The validation is done for all parameters that have both a get and
        set method.

        Arguments:
            verbose: If True, information of checked parameters is printed.

        """
        for k, p in self.parameters.items():
            if hasattr(p, 'get') and hasattr(p, 'set'):
                value = p.get()
                if verbose:
                    print('validate_status: param %s: %s' % (k, value))
                p.validate(value)

    def add_parameter(self, name, parameter_class=Parameter, parent=None, **kwargs):
        """
        Bind one Parameter to this instrument.

        Instrument subclasses can call this repeatedly in their ``__init__``
        for every real parameter of the instrument.

        In this sense, parameters are the state variables of the instrument,
        anything the user can set and/or get

        Args:
            name (str): How the parameter will be stored within
                ``instrument.parameters`` and also how you address it using the
                shortcut methods: ``instrument.set(param_name, value)`` etc.

            parameter_class (Optional[type]): You can construct the parameter
                out of any class. Default ``StandardParameter``.

            **kwargs: constructor arguments for ``parameter_class``.

        Returns:
            Newly created parameter

        Raises:
            KeyError: if this instrument already has a parameter with this
                name.
        """
        if name in self.parameters:
            raise KeyError('Duplicate parameter name {}'.format(name))

        if parent is None:
            parent = self

        param = parameter_class(name=name, parent=parent, **kwargs)
        self.parameters[name] = param

        return param

    def get(self, parameter):
        return self.parameters[parameter].get()
Пример #15
0
 def __init__(self):
     self.timings = DotDict()
    def analyse(self, traces: Dict[str, Dict[str, np.ndarray]] = None, plot: bool = False):
        """Analyse flipping events between nuclear states

        Returns:
            (Dict[str, Any]): Dict containing:

            :results_read (dict): `analyse_traces` results for each read
              trace
            :up_proportions_{idx} (np.ndarray): Up proportions, the
              dimensionality being equal to ``NMRParameter.samples``.
              ``{idx}`` is replaced with the zero-based ESR frequency index.
            :Results from `analyse_flips`. These are:

              - flips_{idx},
              - flip_probability_{idx}
              - combined_flips_{idx1}{idx2}
              - combined_flip_probability_{idx1}{idx2}

              Additionally, each of the above results will have another result
              with the same name, but prepended with ``filtered_``, and appended
              with ``_{idx1}{idx2}`` if not already present. Here, all the
              values are filtered out where the corresponding pair of
              up_proportion samples do not have exactly one high and one low for
              each sample. The values that do not satisfy the filter are set to
              ``np.nan``.

              :filtered_scans_{idx1}{idx2}:
        """
        if traces is None:
            traces = self.traces
        # Convert to DotDict so we can access nested pulse sequences
        traces = DotDict(traces)

        self.results = DotDict()

        initialization_sequence = getattr(self.pulse_sequence, 'initialization', None)
        if initialization_sequence is not None and initialization_sequence.enabled:
            # An initialization sequence is added, we need to filter the results
            # based on whether initialization was successful
            self.results["initialization"] = self.analyses.initialization.analyse(
                traces=traces.ESR_initialization, plot=plot
            )
            try:
                filtered_shots = next(
                    val for key, val in self.results["initialization"].items()
                    if key.startswith("filtered_shots")
                )
            except StopIteration:
                logger.warning(
                    "No filtered_shots found, be sure to set "
                    "analyses.initialization.settings.threshold_up_proportion"
                )
                filtered_shots = self.results.initialization.up_proportions > 0.5
        else:
            # Do not use filtered shots
            filtered_shots = None

        self.results["ESR"] = self.analyses.ESR.analyse(
            traces=traces.ESR, plot=plot
        )

        up_proportions_arrs = np.array([
            val for key, val in self.results['ESR'].items()
            if key.startswith('up_proportion') and 'idxs' not in key
        ])
        if self.analyses.NMR.enabled:
            self.results["NMR"] = self.analyses.NMR.analyse(
                up_proportions_arrs=up_proportions_arrs,
                filtered_shots=filtered_shots
            )
        if self.analyses.NMR_electron_readout.enabled:
            self.results["NMR_electron_readout"] = self.analyses.NMR_electron_readout.analyse(
                traces=traces.NMR, filtered_shots=filtered_shots, plot=plot,
                threshold_voltage=self.results['ESR']['threshold_voltage']
            )

        return self.results