uniform_frame_variations_format#
Uniform mirror of frame_variations_format: UT3Frame + UT3Variations.
Supercores + host-numpy masks instead of ragged core tuples. ut3_orthogonal_representations
returns the frontend objects (the backend ufv_conversions twin returns raw data). Class
asymmetries vs the ragged classes (e.g. no save/to_vector) are deliberate design, not gaps.
Classes#
Frame (orthogonal frame) for the frame-variations representation of uniform Tucker tensor trains. |
|
Variation cores for the frame-variations representation of uniform Tucker tensor trains. |
|
Diagonal weights defining a metric on the tangent coordinates of a |
Functions#
|
Construct frame-variation representations of UniformTuckerTensorTrain with orthogonal frame. |
|
Check that |
|
Absorb the metric |