Struct TensorValue 

pub struct TensorValue { /* private fields */ }

Owned tensor value with one move-only physical owner and metadata-only layout.

TensorValue is intentionally not cloneable. View transformations consume the value and move its existing owner; TensorValue::duplicate is the explicit boundary for creating another physical allocation.

Implementations

impl TensorValue

pub fn duplicate(&self) -> Result<TensorValue, Error>

Explicitly duplicate the physical owner represented by this value.

Errors

Returns [crate::Error::RuntimeState] or [crate::Error::Unsupported] when the backend/storage owner cannot be duplicated.

pub fn from_tensor(tensor: Tensor) -> TensorValue

pub fn from_parts( tensor: Tensor, shape: Vec<usize>, strides: Vec<isize>, offset: isize, ) -> Result<TensorValue, Error>

Errors

Returns [ValidationError::InvalidArgument] or [ValidationError::IntegerOverflow] when the supplied layout is invalid for the tensor’s physical buffer.

pub fn into_tensor(self) -> Result<Tensor, Error>

Consume an unshared compact value and return its physical owner.

Errors

Returns [crate::Error::Unsupported] for a metadata-only view or [crate::Error::RuntimeState] when another value retains the shared owner.

pub fn as_tensor(&self) -> Option<&Tensor>

pub fn is_view(&self) -> bool

pub fn dtype(&self) -> DType

pub fn shape(&self) -> &[usize]

pub fn strides(&self) -> &[isize]

pub fn offset(&self) -> isize

pub fn tensor_view(&self) -> TensorView<'_>

pub fn tensor_read(&self) -> TensorRead<'_>

pub fn transpose_view( self, axes: impl AsRef<[usize]>, ) -> Result<TensorValue, Error>

Errors

Returns [crate::Error::Validation] with [tenferro_tensor_core::ValidationError::InvalidPermutationLength], [tenferro_tensor_core::ValidationError::AxisOutOfBounds], or [tenferro_tensor_core::ValidationError::DuplicateAxis] when axes is not a valid permutation of the value rank.

pub fn try_reshape_view( self, shape: impl IntoShapeVec, ) -> Result<TensorValue, TensorValueViewError>

Errors

Returns [crate::Error::Validation] with [tenferro_tensor_core::ValidationError::NonContiguousViewAsSlice] when the source is not compact column-major, [tenferro_tensor_core::ValidationError::ShapeMismatch] (whose [tenferro_tensor_core::ShapeMismatch::ReshapeElementCount] source records the counts) when element counts differ, [tenferro_tensor_core::ValidationError::IntegerOverflow] for shape arithmetic overflow, or [tenferro_tensor_core::ValidationError::ViewOutOfBounds] when the reshaped view exceeds the backing buffer.

pub fn reshape_view( self, shape: impl IntoShapeVec, ) -> Result<TensorValue, Error>

Errors

Returns [crate::Error::Validation] with [tenferro_tensor_core::ValidationError::NonContiguousViewAsSlice] when the source is not compact, [tenferro_tensor_core::ValidationError::ShapeMismatch] when element counts differ, or [tenferro_tensor_core::ValidationError::IntegerOverflow] / [tenferro_tensor_core::ValidationError::ViewOutOfBounds] for invalid target-shape arithmetic or bounds.

pub fn slice_view(self, config: &SliceConfig) -> Result<TensorValue, Error>

Errors

Returns [crate::Error::Validation] with [tenferro_tensor_core::ValidationError::RankMismatch] when a slice vector does not match the value rank, [tenferro_tensor_core::ValidationError::InvalidArgument] when a bound or stride cannot be represented or is invalid, [tenferro_tensor_core::ValidationError::InvalidSliceStep] or [tenferro_tensor_core::ValidationError::InvalidSliceBounds] for slice parameters, [tenferro_tensor_core::ValidationError::IntegerOverflow] for slice arithmetic overflow, or [tenferro_tensor_core::ValidationError::ViewOutOfBounds] when the result exceeds the backing buffer.

pub fn broadcast_in_dim_view( self, shape: impl IntoShapeVec, dims: impl AsRef<[usize]>, ) -> Result<TensorValue, Error>

Errors

Returns [crate::Error::Validation] with [tenferro_tensor_core::ValidationError::RankMismatch], [tenferro_tensor_core::ValidationError::AxisOutOfBounds], or [tenferro_tensor_core::ValidationError::DuplicateAxis] for invalid dimension mappings, [tenferro_tensor_core::ValidationError::ShapeDataLengthMismatch] for incompatible extents, [tenferro_tensor_core::ValidationError::ViewOutOfBounds] when the result exceeds the backing buffer, or [tenferro_tensor_core::ValidationError::IntegerOverflow] for layout arithmetic overflow.

Trait Implementations

impl Debug for TensorValue

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

Formats the value using the given formatter.