Function dfdx::tensor_ops::normalize_axis
source · [−]pub fn normalize_axis<T, const I: isize>(t: T, epsilon: T::Dtype) -> T where
T: Reduce1<I>,
T::Array: HasAxis<I>,
Expand description
Normalizes t
to have mean 0.0
and stddev 1.0
along the I
dimension of T
. epsilon
is passed to std_axis().
Computes (t - t.mean(I)) / t.std(I, epsilon)
.
Related functions: mean_axis(), std_axis(), var_axis()
Examples
let a = Tensor1D::new([-2.0, -1.0, 0.0, 5.0, 2.0]);
let r = a.normalize_axis::<-1>(1e-5);
assert!(r.clone().mean_axis::<-1>().data().abs() < 1e-6);
assert!((r.clone().std_axis::<-1>(0.0).data() - 1.0).abs() < 1e-6);