pub struct NormalizeOptions<'a> {
pub pre_scale_bias: f32,
pub scale: f32,
pub element_scale: Option<&'a [f32]>,
pub bias: f32,
pub element_bias: Option<&'a [f32]>,
}
Expand description
Configuration for the Normalize
operation.
Fields§
§pre_scale_bias: f32
Bias to subtract before scaling. This is usually the mean of the data.
scale: f32
Constant scale to multiply each element by. This is usually the inverse standard deviation of the data.
element_scale: Option<&'a [f32]>
Per-element scale to multiply each element by.
bias: f32
Constant bias to add after scaling
element_bias: Option<&'a [f32]>
Per-element bias to add after scaling
Trait Implementations§
Auto Trait Implementations§
impl<'a> Freeze for NormalizeOptions<'a>
impl<'a> RefUnwindSafe for NormalizeOptions<'a>
impl<'a> Send for NormalizeOptions<'a>
impl<'a> Sync for NormalizeOptions<'a>
impl<'a> Unpin for NormalizeOptions<'a>
impl<'a> UnwindSafe for NormalizeOptions<'a>
Blanket Implementations§
Source§impl<T> BorrowMut<T> for Twhere
T: ?Sized,
impl<T> BorrowMut<T> for Twhere
T: ?Sized,
Source§fn borrow_mut(&mut self) -> &mut T
fn borrow_mut(&mut self) -> &mut T
Mutably borrows from an owned value. Read more