bartorch.prox.LocallyLowRank#
- class bartorch.prox.LocallyLowRank(axes, weight, joint_axes=(), *, block=8, randshift=True, overlapping=False)#
Nuclear norm of blocks over
axes(pics -R L).- Parameters:
axes (int or tuple of int) – Axes the blocks span, as indices into the image’s shape.
weight (float)
joint_axes (int or tuple of int, optional) – Axes forming the columns of each block’s matrix.
block (int) – Block edge length (
pics -b).randshift (bool) – Shift the block grid by a random offset;
pics -nturns it off.overlapping (bool) – Fully overlapping blocks instead of a shifted grid (
pics -N).
- __init__(axes, weight, joint_axes=(), *, block=8, randshift=True, overlapping=False)#
Methods
__init__(axes, weight[, joint_axes, block, ...])apply_transform(x[, image_shape, mode])This term's transform applied to
x, without making an operator of it.build(shape)The BART operator for this term over an image of C-order
shape.prox(x[, gamma, image_shape])prox_{gamma f}(x), the operator BART's solvers call.prox_shape(image_shape)The C-order shape this term's proximal operator works on.
rewind(image_shape)Put this term's own random generator back to where it started.
transform(image_shape)The operator BART puts in front of this term's proximal operator.
transform_is_identity(image_shape)Whether that transform is the identity, as BART decides it.
Attributes
axescountjoint_axeskindweight