pub struct DynamicRangeQuantizeDescriptor {
pub n: i32,
pub d: i32,
pub q_min: i32,
pub q_max: i32,
pub mode: DynamicRangeMode,
pub scope: DynamicRangeScope,
pub input_element: ElementKind,
pub output_element: ElementKind,
}Expand description
Descriptor for a dynamic_range_quantize op.
Fields§
§n: i32Number of token rows (first axis of input / output).
d: i32Feature dim (second axis).
q_min: i32Quantization range lower bound (e.g. -127 for symmetric s8).
q_max: i32Quantization range upper bound (e.g. 127 for symmetric s8).
mode: DynamicRangeModeSymmetric / asymmetric mode.
scope: DynamicRangeScopePer-tensor / per-channel / per-token / per-group scope.
input_element: ElementKindInput FP element kind. Must match TIn::KIND.
output_element: ElementKindOutput int element kind (must match TOut::KIND). S8 only.
Trait Implementations§
Source§impl Clone for DynamicRangeQuantizeDescriptor
impl Clone for DynamicRangeQuantizeDescriptor
Source§fn clone(&self) -> DynamicRangeQuantizeDescriptor
fn clone(&self) -> DynamicRangeQuantizeDescriptor
Returns a duplicate of the value. Read more
1.0.0 (const: unstable) · Source§fn clone_from(&mut self, source: &Self)
fn clone_from(&mut self, source: &Self)
Performs copy-assignment from
source. Read moreimpl Copy for DynamicRangeQuantizeDescriptor
Auto Trait Implementations§
impl Freeze for DynamicRangeQuantizeDescriptor
impl RefUnwindSafe for DynamicRangeQuantizeDescriptor
impl Send for DynamicRangeQuantizeDescriptor
impl Sync for DynamicRangeQuantizeDescriptor
impl Unpin for DynamicRangeQuantizeDescriptor
impl UnsafeUnpin for DynamicRangeQuantizeDescriptor
impl UnwindSafe for DynamicRangeQuantizeDescriptor
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