Struct FlexQTensor 

pub struct FlexQTensor { /* private fields */ }

Quantized tensor for the Flex backend.

Stores quantized i8 values in the tensor and keeps scales separately for efficient dequantization without reparsing bytes.

Implementations

impl FlexQTensor

pub fn new( tensor: FlexTensor, scheme: QuantScheme, scales: Vec<f32>, global: Option<f32>, ) -> FlexQTensor

Create a new quantized tensor.

The tensor must store i8 data and scales must be non-empty. A two-level scheme takes a per-tensor scale, and no other level does.

pub fn tensor(&self) -> &FlexTensor

Get the underlying tensor.

pub fn scales(&self) -> &[f32]

Get the quantization scales.

Trait Implementations

impl Clone for FlexQTensor

fn clone(&self) -> FlexQTensor

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for FlexQTensor

fn fmt(&self, f: &mut Formatter<'_>) -> Result<(), Error>

Formats the value using the given formatter.

impl TensorMetadata for FlexQTensor

type Device = FlexDevice

The device type associated with the tensor.

fn dtype(&self) -> DType

Get the dtype of the tensor.

fn shape(&self) -> Shape

Get the shape of the tensor.

fn rank(&self) -> usize

Get the number of dimensions of the tensor.

fn device(&self) -> <FlexQTensor as TensorMetadata>::Device

Get the device associated with the tensor.

fn can_mut(&self) -> bool

Whether the tensor’s buffer can be mutated in place — i.e. this handle uniquely owns it, so an in-place op (slice_assign, an inplace kernel) writes the existing allocation instead of copying it first.

fn scheme(&self) -> QuantScheme

Get the quantization scheme for a quantized float tensor.