Struct Tensor

pub struct Tensor {
    pub dimensions: Dimensions,
    pub value_types: ValueTypes,
    pub flags: u8,
    pub data: Vec<u8>,
}

The tensor’s dimensions and type are provided as metadata to a model. Any metadata shall be associated to the respective model in a blob store.

Fields

dimensions: Dimensions

Tensor Dimensions The Dimension array contains one size value for each dimension of the Tensor

value_types: ValueTypes

The types array contains either: a single ValueType that represents the data values for all dimensions (homogeneous array) or one ValueType per dimension. In other words, the length of this array is either 1 or the length of dimensions.

flags: u8

Optional bit flags representing the data representation in the Tensor. Currently only one bit (LSB) is used to indicate row-major order (0) or column-major order (1).

data: Vec<u8>

The Tensor

Implementations

impl Tensor

pub fn shape(&self) -> Vec<usize>

Returns dimensions as an array of usize

pub fn set_row_major(&mut self)

Sets the tensor flag to indicate row-major order Setting this flag does not alter the data array

pub fn is_row_major(&self) -> bool

Returns true if the tensor is row-major order (based on flags setting)

pub fn set_column_major(&mut self)

Sets the tensor flag to indicate column-major order Setting this flag does not alter the data array

pub fn is_column_major(&self) -> bool

Returns true if the tensor is column-major order (based on flags setting)

pub fn set_little_endian(&mut self)

Sets the tensor flag to indicate data is in big endian format

pub fn is_little_endian(&self) -> bool

Returns true if the tensor flags indicate that data is little endian

pub fn set_big_endian(&mut self)

Sets the tensor flag to indicate data is in big endian format

pub fn is_big_endian(&self) -> bool

Returns true if the tensor flags indicate that data is big endian

pub fn check_dims(&self) -> Result<(), String>

perform validation checking on dimensions

Trait Implementations

impl Clone for Tensor

fn clone(&self) -> Tensor

Returns a copy of the value.

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

Performs copy-assignment from source.

impl Debug for Tensor

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

Formats the value using the given formatter.

impl Default for Tensor

fn default() -> Tensor

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for Tensor

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl PartialEq for Tensor

fn eq(&self, other: &Tensor) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Serialize for Tensor

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl Eq for Tensor

impl StructuralPartialEq for Tensor