decompress()

BasisMulti.decompress(states_compressed, locs=None)

Map low-dimensional latent coordinates to high-dimensional states.

Parameters:

states_compressed(r, …) ndarray

Matrix of r-dimensional latent coordinate vectors, or a single coordinate vector. The first reduced_variable_sizes[0] entries correspond to the first reduced state variable, the next reduced_variable_sizes[1] entries correspond to the second reduced state variable, and so on.

locsslice or (p,) ndarray of integers or None

If given, decompress each state variable at only the p specified locations (indices) described by locs. This option requires each full state variable to have the same dimension.

Returns:

states_decompressed(n, …) or (num_variables*p, …) ndarray

Matrix of n-dimensional decompressed state vectors, or the \(n_q \cdot p\) entries of such at the entries specified by locs.