fv_conversions

Raw-data (frame, variations) conversions for the ragged layer.

t3_orthogonal_representations factors t3 data into the orthogonal frame + gauged variations (data-level twin of the frontend function); fv_to_t3 reconstructs the SINGLE term selected by a frame/variations pairing – not the tangent sum (that is tv_operations.tv_to_t3; same-looking near-twins, different math).

Functions

fv_to_t3(index, frame, variations)	Convert ith frame-variation representation to TuckerTensorTrain.
t3_orthogonal_representations(x[, ...])	Construct frame-variation representations of TuckerTensorTrain with orthogonal frame.