pub struct QuantizationInfo {
pub scale: TensorHandle<TestRuntime>,
pub scheme: QuantScheme,
pub shape: Shape,
}Expand description
Information about a quantized tensor in tests. This allows marking a tensor as quantized for the kernel dispatcher while keeping the original unquantized data on the host for reference.
Fields§
§scale: TensorHandle<TestRuntime>The scale tensor on the device.
scheme: QuantSchemeThe quantization scheme (e.g., Symmetric, Tensor-wise, etc.)
shape: ShapeThe original unquantized shape of the tensor.
Trait Implementations§
Auto Trait Implementations§
impl !RefUnwindSafe for QuantizationInfo
impl !UnwindSafe for QuantizationInfo
impl Freeze for QuantizationInfo
impl Send for QuantizationInfo
impl Sync for QuantizationInfo
impl Unpin for QuantizationInfo
impl UnsafeUnpin for QuantizationInfo
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