ForEachBlockPass

class ForEachBlockPass(loop_body, calculate_error_bound=False, collection_filter=None, replace_filter='always', batch_size=None)[source]

Bases: BasePass

A pass that executes other passes on each block in the circuit.

This is a control pass that executes a workflow on every block in the circuit. This will be done in parallel.

__init__(loop_body, calculate_error_bound=False, collection_filter=None, replace_filter='always', batch_size=None)[source]

Construct a ForEachBlockPass.

Parameters:

  • loop_body (WorkflowLike) – The workflow to execute on every block.

  • calculate_error_bound (bool) – If set to true, will calculate errors on blocks after running loop_body on them and use these block errors to calculate an upper bound on the full circuit error. (Default: False)

  • collection_filter (Callable[[Operation], bool] | None) – A predicate that determines which operations should have loop_body called on them. Called with each operation in the circuit. If this returns true, that operation will be formed into an individual circuit and passed through loop_body. Defaults to all CircuitGates, ConstantUnitaryGates, and VariableUnitaryGates. #TODO: address importability

  • replace_filter (ReplaceFilterFn | str | None) –

    A predicate that determines if the resulting circuit, after calling loop_body on a block, should replace the original operation. Called with the circuit output from loop_body and the original operation. If this returns true, the operation will be replaced with the new circuit. Defaults to always replace. If none is passed, will generate a replace filter always replaces. If a string is passed, will generate a replace filter corresponding to the string. The string should either be ‘always’, ‘less-than’, ‘less-than-multi’, ‘less-than-many’, ‘less-than-respecting’, ‘less-than-respecting-multi’, or ‘less-than-respecting-many’.

    • ’always’ will always replace

    • ’less-than’ will replace if the new circuit has fewer

      gates than the old circuit.

    • ’less-than-multi’ will replace if the new circuit has

      fewer multi-qudit gates than the old circuit.

    • ’less-than-many’ will replace if the new circuit has

      fewer many-qudit gates than the old circuit.

    • ’less-than-respecting’ will replace if the new circuit

      has fewer gates than the old circuit or the old doesn’t respect the model (ignoring single-qudit gate sets).

    • ’less-than-respecting-multi’ will replace if the new

      circuit has fewer multi-qudit gates than the old circuit or the old doesn’t respect the model (ignoring single-qudit gate sets).

    • ’less-than-respecting-many’ will replace if the new

      circuit has fewer many-qudit gates than the old circuit or the old doesn’t respect the model (ignoring single-qudit gate sets).

    • ’less-than-respecting-fully’ will replace if the new

      circuit has fewer gates than the old circuit or the old doesn’t respect the model.

    • ’less-than-respecting-fully-multi’ will replace if

      the new circuit has fewer multi-qudit gates than the old circuit or the old doesn’t respect the model.

    • ’less-than-respecting-fully-many’ will replace if

      the new circuit has fewer many-qudit gates than the old circuit or the old doesn’t respect the model.

    Defaults to ‘always’. #TODO: address importability

  • batch_size (int) – (Deprecated).

Attributes

key

The key in data, where block data will be put.

name

The name of the pass.

pass_down_block_specific_key_prefix

Data specific to the processing of individual blocks in a partitioned circuit can be injected into the PassData in run by using this prefix.

pass_down_key_prefix

If a key exists in the pass data with this prefix, pass it to blocks.

Methods

execute(*args, **kwargs)

Map a function over iterable arguments in parallel.

get_connectivity(_, data)

Retrieve the current connectivity of the circuit.

get_model(_, data)

Retrieve the machine model from the data dictionary.

get_placement(_, data)

Retrieve the logical to physical qubit map from the data dictionary.

get_target(_, data)

Retrieve the target from the data dictionary.

in_parallel(data)

Return true if pass is being executed in a parallel.

run(circuit, data)

Perform the pass's operation, see BasePass for more.