pyhgf.typing.typing.LearningSequence#

class pyhgf.typing.typing.LearningSequence(prediction_steps, update_steps, learning_steps)[source]#

Set of update functions to update the weights of a deep network.

Parameters:

  • prediction_steps (tuple[tuple[int, jaxlib.xla_extension.PjitFunction], ...]) – The sequence of prediction steps applied top-down before observation.

  • update_steps (tuple[tuple[int, jaxlib.xla_extension.PjitFunction], ...]) – The sequence of posterior and prediction-error update steps.

  • learning_steps (tuple[tuple[int, jaxlib.xla_extension.PjitFunction], ...]) – The sequence of weight-learning steps applied after the update steps.

__init__(*args, **kwargs)#

Methods

__init__(*args, **kwargs)

count(value, /)

Return number of occurrences of value.

index(value[, start, stop])

Return first index of value.

Attributes

learning_steps

Alias for field number 2

prediction_steps

Alias for field number 0

update_steps

Alias for field number 1