neuraxle.steps.loop

Module-level documentation for neuraxle.steps.loop. Here is an inheritance diagram, including dependencies to other base modules of Neuraxle:

Pipeline Steps For Looping

Classes

Break()
BreakIf(condition_function)
Continue()
ContinueIf(condition_function)
FlattenForEach(wrapped, then_unflatten)	Step that reduces a dimension instead of manually looping on it.
ForEach(wrapped[, cache_folder_when_no_handle])	Truncable step that fits/transforms each step for each of the data inputs, and expected outputs.
StepClonerForEachDataInput(wrapped[, copy_op])

Examples using neuraxle.steps.loop.FlattenForEach

Create label encoder across multiple columns

Examples using neuraxle.steps.loop.ForEach

Parallel processing in Neuraxle

Exceptions

BreakInterrupt	This exception is used to signal the interruption of a minibatch
ContinueInterrupt	This exception is used to signal to the minibatch iterator to skip the rest of the execution of the current iteration.

class neuraxle.steps.loop.ForEach(wrapped: neuraxle.base.BaseTransformer, cache_folder_when_no_handle=None)

Bases: neuraxle.base.ForceHandleOnlyMixin, neuraxle.base.MetaStep

Truncable step that fits/transforms each step for each of the data inputs, and expected outputs.

See also

BaseStep, BaseSaver, BaseHasher, NonTransformableMixin, Pipeline, HyperparameterSamples, HyperparameterSpace, DataContainer

__init__(wrapped: neuraxle.base.BaseTransformer, cache_folder_when_no_handle=None)

Initialize self. See help(type(self)) for accurate signature.

_fit_data_container(data_container: neuraxle.data_container.DataContainer, context: neuraxle.base.ExecutionContext) → neuraxle.base.BaseStep

Fit each step for each data inputs, and expected outputs

Return type

BaseStep

Parameters

• data_container (DataContainer) – data container

• context (ExecutionContext) – execution context

Returns

self

_transform_data_container(data_container: neuraxle.data_container.DataContainer, context: neuraxle.base.ExecutionContext) → neuraxle.data_container.DataContainer

Transform each step for each data inputs.

Return type

DataContainer

Parameters

• data_container (DataContainer) – data container

• context (ExecutionContext) – execution context

Returns

self

_fit_transform_data_container(data_container: neuraxle.data_container.DataContainer, context: neuraxle.base.ExecutionContext) → Tuple[neuraxle.base.BaseStep, neuraxle.data_container.DataContainer]

Fit transform each step for each data inputs, and expected outputs

Parameters

• data_container (DataContainer) – data container to fit transform

• context (ExecutionContext) – execution context

Returns

self, transformed_data_container

_abc_impl = <_abc_data object>

exception neuraxle.steps.loop.ContinueInterrupt

Bases: Exception

This exception is used to signal to the minibatch iterator to skip the rest of the execution of the current iteration.

exception neuraxle.steps.loop.BreakInterrupt

Bases: Exception

This exception is used to signal the interruption of a minibatch

class neuraxle.steps.loop.Break

Bases: neuraxle.base.ForceHandleMixin, neuraxle.base.Identity

__init__()

Initialize self. See help(type(self)) for accurate signature.

_did_process(data_container: neuraxle.data_container.DataContainer, context: neuraxle.base.ExecutionContext) → neuraxle.data_container.DataContainer

Apply side effects after any step method.

Return type

DataContainer

Parameters

• data_container (DataContainer) – data container

• context (ExecutionContext) – execution context

Returns

(data container, execution context)

_abc_impl = <_abc_data object>

class neuraxle.steps.loop.BreakIf(condition_function: Callable)

Bases: neuraxle.steps.flow.ExecuteIf

__init__(condition_function: Callable)

Initialize self. See help(type(self)) for accurate signature.

_abc_impl = <_abc_data object>

class neuraxle.steps.loop.Continue

Bases: neuraxle.base.ForceHandleMixin, neuraxle.base.Identity

__init__()

Initialize self. See help(type(self)) for accurate signature.

_did_process(data_container: neuraxle.data_container.DataContainer, context: neuraxle.base.ExecutionContext) → neuraxle.data_container.DataContainer

Apply side effects after any step method.

Return type

DataContainer

Parameters

• data_container (DataContainer) – data container

• context (ExecutionContext) – execution context

Returns

(data container, execution context)

_abc_impl = <_abc_data object>

class neuraxle.steps.loop.ContinueIf(condition_function: Callable)

Bases: neuraxle.steps.flow.ExecuteIf

__init__(condition_function: Callable)

Initialize self. See help(type(self)) for accurate signature.

_abc_impl = <_abc_data object>

class neuraxle.steps.loop.StepClonerForEachDataInput(wrapped: neuraxle.base.BaseTransformer, copy_op=<function deepcopy>)

Bases: neuraxle.base.ForceHandleOnlyMixin, neuraxle.base.MetaStep

__init__(wrapped: neuraxle.base.BaseTransformer, copy_op=<function deepcopy>)

Initialize self. See help(type(self)) for accurate signature.

get_children() → List[neuraxle.base.BaseStep]

Get the list of all the children for that step.

Returns

list of children. The first is the original wrapped step, the others are the steps that are cloned.

_will_process(data_container: neuraxle.data_container.DataContainer, context: neuraxle.base.ExecutionContext) → Tuple[neuraxle.base.BaseStep, neuraxle.data_container.DataContainer]

Apply side effects before any step method. :type context: ExecutionContext :type data_container: DataContainer :param data_container: data container :param context: execution context :return: (data container, execution context)

_copy_one_step_per_data_input(data_container)

_fit_transform_data_container(data_container: neuraxle.data_container.DataContainer, context: neuraxle.base.ExecutionContext) → Tuple[neuraxle.base.BaseStep, neuraxle.data_container.DataContainer]

Fit and transform data container with the given execution context. Will do:

data_container, context = self._fit_data_container(data_container, context)
data_container = self._transform_data_container(data_container, context)
return self, data_container

Parameters

• data_container (DataContainer) – data container

• context (ExecutionContext) – execution context

Returns

transformed data container

_fit_data_container(data_container: neuraxle.data_container.DataContainer, context: neuraxle.base.ExecutionContext) → Tuple[neuraxle.base.BaseStep, neuraxle.data_container.DataContainer]

Fit data container.

Parameters

• data_container (DataContainer) – data container

• context (ExecutionContext) – execution context

Returns

(fitted self, data container)

_transform_data_container(data_container: neuraxle.data_container.DataContainer, context: neuraxle.base.ExecutionContext) → Tuple[neuraxle.base.BaseStep, neuraxle.data_container.DataContainer]

Transform data container.

Parameters

• data_container (DataContainer) – data container

• context (ExecutionContext) – execution context

Returns

data container

_inverse_transform_data_container(data_container: neuraxle.data_container.DataContainer, context: neuraxle.base.ExecutionContext) → neuraxle.data_container.DataContainer

inverse_transform(processed_outputs)

Inverse Transform the given transformed data inputs.

p = Pipeline([MultiplyByN(2)])
_in = np.array([1, 2])
_out = p.transform(_in)
_regenerated_in = p.inverse_transform(_out)
assert np.array_equal(_regenerated_in, _in)
assert np.array_equal(_out, _in * 2)

Parameters

processed_outputs – processed data inputs

Returns

inverse transformed processed outputs

items()

values()

keys()

_abc_impl = <_abc_data object>

class neuraxle.steps.loop.FlattenForEach(wrapped: neuraxle.base.BaseTransformer, then_unflatten: bool = True)

Bases: neuraxle.base.ForceHandleMixin, neuraxle.base.MetaStep

Step that reduces a dimension instead of manually looping on it.

See also

BaseStep, BaseSaver, BaseHasher, MetaStepMixin, NonTransformableMixin, Pipeline, HyperparameterSamples, HyperparameterSpace, DataContainer

__init__(wrapped: neuraxle.base.BaseTransformer, then_unflatten: bool = True)

Initialize self. See help(type(self)) for accurate signature.

_will_process(data_container: neuraxle.data_container.DataContainer, context: neuraxle.base.ExecutionContext) → Tuple[neuraxle.base.BaseStep, neuraxle.data_container.DataContainer]

Flatten data container before any processing is done on the wrapped step.

Parameters

• data_container (DataContainer) – data container to flatten

• context (ExecutionContext) – execution context

Returns

(data container, execution context)

_flatten_list(list_to_flatten)

Flatten the first dimension of a list.

Parameters

list_to_flatten – list to flatten

Returns

flattened list, len flattened lists

_did_process(data_container: neuraxle.data_container.DataContainer, context: neuraxle.base.ExecutionContext) → neuraxle.data_container.DataContainer

Reaugment the flattened data container.

Return type

DataContainer

Parameters

• data_container (DataContainer) – data container to then_unflatten

• context (ExecutionContext) – execution context

Returns

data container

_reaugment_list(list_to_reaugment, flattened_dimension_lengths)

Reaugment list with the flattened dimension lengths.

Parameters

list_to_reaugment – list to then_unflatten

Returns

reaugmented numpy array

_abc_impl = <_abc_data object>