torch.distributions.transforms.AffineTransform
class AffineTransform extends Transformnew AffineTransform(loc: number | Tensor, scale: number | Tensor, options?: AffineTransformOptions)
- readonly
domain(unknown) - readonly
codomain(unknown) - readonly
bijective(unknown) - readonly
loc(Tensor) - – Location parameter.
- readonly
scale(Tensor) - – Scale parameter.
- readonly
event_dim(number) - – Number of event dimensions this transform operates over.
- readonly
sign(number) - – Sign of the Jacobian determinant.
Affine transform: y = loc + scale * x (location-scale transformation).
Transform via the pointwise affine mapping y = loc + scale * x. This is one of the most commonly used transforms, allowing location-scale families to be built from standard distributions.