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)