aspcore.adaptivefilter.LMS

class aspcore.adaptivefilter.LMS(ir_len, num_in, num_out, step_size, regularization, normalization='channel_independent', wait_until_initialized=False, filter_type=None)

Bases: AdaptiveFilterBase

Sample by sample processing, although it accepts block inputs - Dimension of filter is (input channels, output channels, IR length)

  • ir_len, num_in, num_out are integers

  • step_size is either a real positive scalar, or a generator that

    outputs a real positive scalar for each sample

  • regularization is a real positive scalar

  • normalization is a string corresponding to one of the normalization options

  • If wait_until_initialized is True, the ir will not be updated until

    the full buffer of reference signals is filled.

  • filter_type is deprecated and will be removed later.

__init__(ir_len, num_in, num_out, step_size, regularization, normalization='channel_independent', wait_until_initialized=False, filter_type=None)

Methods

__init__(ir_len, num_in, num_out, step_size, ...)

insert_in_signal(ref)

prepare()

process(signalToProcess)

reinitialize()

set_ir(new_ir)

update(ref, desired)

Inputs should be of the shape (channels, num_samples)

Attributes

ir

update(ref, desired)

Inputs should be of the shape (channels, num_samples)