TorchVinecop.inverse_rosenblatt

TorchVinecop.inverse_rosenblatt(u, *, num_threads=1, x=None, batched=None)

Inverse Rosenblatt transform: independent uniforms to dependent uniforms.

Parameters:

uarray, shape (n, d), dtype float: Independent uniforms in [0, 1]^d.
num_threadsint, default=1: Accepted for parity; ignored.
xarray, shape (n, p), or None, optional: External covariates (non-simplified vines), else None.
batchedbool or None, optional: Must be False / None. The inverse cascade has cross-tree dependencies that the per-tree-level wavefront cannot satisfy, so batched=True raises NotImplementedError.

Returns:

array, shape (n, d), dtype float: Dependent uniforms in [1e-10, 1 - 1e-10].

Raises: