Create Pipeline Steps that require implementing only handler methods

If a pipeline step only needs to implement handler methods, then you can inherit from the ForceHandleMixin as demonstrated here. Handler methods are useful when :

• You need to change the shape of the data container passed to the following steps, or the wrapped steps.

• You want to apply side effects based on the data container, and the execution context.

• You want to change the pipeline execution flow based on the data container, and the execution context.

from typing import Tuple

import numpy as np
from neuraxle.base import DACT, BaseStep
from neuraxle.base import ExecutionContext as CX
from neuraxle.base import ForceHandleMixin


class ForceHandleMixinStep(ForceHandleMixin, BaseStep):
    """
    Please make your steps inherit from ForceHandleMixin when they only implement handle_methods, but also
    when you want to make impossible the use of regular fit, transform, and fit_transform methods
    Also, make sure that BaseStep is the last step you inherit from.
    """

    def __init__(self):
        BaseStep.__init__(self)
        ForceHandleMixin.__init__(self)

    def _fit_data_container(self, data_container: DACT, context: CX) -> BaseStep:
        """
        Change the shape of the data container.
        and/or
        Apply any side effects based on the data container
        And/or
        Change the execution flow of the pipeline
        """
        context.logger.info("Handling the 'fit' with handler method!")
        return self

    def _transform_data_container(self, data_container: DACT, context: CX) -> DACT:
        """
        Change the shape of the data container.
        and/or
        Apply any side effects based on the data container
        And/or
        Change the execution flow of the pipeline
        """
        context.logger.info("Handling the 'transform' with handler method!")
        return data_container

    def _fit_transform_data_container(
        self, data_container: DACT, context: CX
    ) -> Tuple[BaseStep, DACT]:
        """
        Change the shape of the data container.
        and/or
        Apply any side effects based on the data container
        And/or
        Change the execution flow of the pipeline
        """
        context.logger.info("Handling the 'fit_transform' with handler method!")
        return self, data_container


def main():
    p = ForceHandleMixinStep()
    data_inputs = np.array([0, 1])
    expected_outputs = np.array([0, 1])

    p = p.fit(data_inputs, expected_outputs)
    outputs = p.transform(data_inputs)


if __name__ == '__main__':
    main()