Struct QuantizedVector 

pub struct QuantizedVector {
    pub data: QuantizedData1D,
    pub length: usize,
    pub data_type: QuantizedDataType,
}

A vector with quantized values

Fields

data: QuantizedData1D

The quantized values can be stored in different formats

length: usize

The original length of the vector

data_type: QuantizedDataType

The data type used for quantization

Implementations

impl QuantizedVector

pub fn new_i8( data: Array1<i8>, length: usize, datatype: QuantizedDataType, ) -> Self

Creates a new quantized vector with int8 storage

pub fn new_f16(data: Array1<f16>, length: usize) -> Self

Creates a new f16 quantized vector

pub fn new_bf16(data: Array1<bf16>, length: usize) -> Self

Creates a new bf16 quantized vector

pub fn from_i8(data: Array1<i8>, length: usize) -> Self

Creates a standard Int8 quantized vector (for backward compatibility)

pub fn get(&self, idx: usize) -> i8

👎Deprecated since 0.1.0:

Use get_i8 or get_f32 instead

pub fn len(&self) -> usize

Returns the length of the vector

pub fn is_empty(&self) -> bool

Returns true if the vector is empty

pub fn get_i8(&self, idx: usize) -> i8

Get value at specified position as i8 (for int quantization)

pub fn get_f32(&self, idx: usize) -> f32

Get value at specified position as f32 (for all quantization types)

Trait Implementations

impl Clone for QuantizedVector

fn clone(&self) -> QuantizedVector

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for QuantizedVector

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

Formats the value using the given formatter.