sigpy.convolve_data_adjoint

sigpy.convolve_data_adjoint(output, filt, data_shape, mode='full', strides=None, multi_channel=False)[source]

Adjoint convolution operation with respect to data.

Note that the cuDNN version only supports inputs with D=1, 2 or 3.

Parameters:
  • output (array) – output array of shape \([..., p_1, ..., p_D]\) if multi_channel is False, \([..., c_o, p_1, ..., p_D]\) otherwise.
  • filt (array) – filter array of shape \([n_1, ..., n_D]\) if multi_channel is False \([c_o, c_i, n_1, ..., n_D]\) otherwise.
  • mode (str) – {‘full’, ‘valid’}.
  • strides (None or tuple of ints) – convolution strides of length D.
  • multi_channel (bool) – specify if input/output has multiple channels.
  • multi_channel – specify if data/output has multiple channels.
  • mode – {‘full’, ‘valid’}.
Returns:

data array of shape

\([..., m_1, ..., m_D]\) if multi_channel is False, \([..., c_i, m_1, ..., m_D]\) otherwise.

Return type:

array