[−][src]Struct autograd::tensor::Tensor

pub struct Tensor<T: Float>(pub Rc<TensorCore<T>>);

Symbolic multi-dimensional array.

Methods

`impl<T: Float> Tensor<T>`[src]

`pub fn get_persistent_array(&self) -> Option<&NdArray<T>>`[src]

Returns a reference to the persistent array.

Returns Some if this tensor is made from ag::variable or ag::constant.

`pub unsafe fn get_persistent_array_mut(&self) -> Option<&mut NdArray<T>>`[src]

Returns a mutable reference to the persistent array.

Returns Some if this tensor is made from ag::variable.

`pub fn has_persistent_array(&self) -> bool`[src]

Returns True if this tensor is made from ag::variable or ag::constant.

`impl<'a, 'b: 'a, 'c: 'a, T: Float + 'b + 'c> Tensor<T>`[src]

`pub fn builder() -> TensorBuilder<T>`[src]

`pub fn eval<I>(&self, feeds: I) -> Option<NdArray<T>> where I: IntoIterator<Item = &'a (&'b Tensor<T>, &'c Array<T, IxDyn>)>,` [src]

Evaluates this tensor as an ndarray's array object.

See eval.

`pub fn shape(&self) -> Tensor<T>`[src]

Returns the (symbolic) shape of this tensor.

See shape.

`pub fn rank(&self) -> Tensor<T>`[src]

Returns the (symbolic) rank of this tensor.

See rank.

`pub fn size(&self) -> Tensor<T>`[src]

Returns the (symbolic) size of this tensor.

See size.

`impl<T: Float> Tensor<T>`[src]

`pub fn get(&self, i: isize) -> Tensor<T>`[src]

Looks up a symbolic element from this tensor.

Index i can be negative.

extern crate ndarray;
extern crate autograd as ag;

let ref a = ag::variable(ndarray::arr2(&[[2., 3.], [4., 5.]]));
let ref b = a.get(2);

assert_eq!(b.eval(&[]).unwrap()[ndarray::IxDyn(&[])], 4.);

Methods from Deref<Target = Rc<TensorCore<T>>>

Trait Implementations

`impl<T: Float> ArrayLike<T> for Tensor<T>`[src]

`fn as_tensor(&self) -> Tensor<T>`[src]

`impl<T: Float> Clone for Tensor<T>`[src]

`fn clone(&self) -> Tensor<T>`[src]

`default fn clone_from(&mut self, source: &Self)`
1.0.0
[src]

Performs copy-assignment from source. Read more

`impl<'a, T: Float> Ord for &'a Tensor<T>`[src]

`fn cmp(&self, other: &&'a Tensor<T>) -> Ordering`[src]

Compares the addresses of the two tensors. This can be used for ordering-based data structures (e.g. BinaryTree).

`default fn max(self, other: Self) -> Self`
1.21.0
[src]

Compares and returns the maximum of two values. Read more

`default fn min(self, other: Self) -> Self`
1.21.0
[src]

Compares and returns the minimum of two values. Read more

`default fn clamp(self, min: Self, max: Self) -> Self`[src]

🔬 This is a nightly-only experimental API. (clamp)

Restrict a value to a certain interval. Read more

`impl<T: Float> Eq for Tensor<T>`[src]

`impl<'a, T: Float> PartialOrd<&'a Tensor<T>> for &'a Tensor<T>`[src]

`fn partial_cmp(&self, other: &&'a Tensor<T>) -> Option<Ordering>`[src]

Compares the addresses of the two tensors. This can be used for ordering-based data structures (e.g. BinaryTree).

`#[must_use] default fn lt(&self, other: &Rhs) -> bool`
1.0.0
[src]

This method tests less than (for self and other) and is used by the < operator. Read more

`#[must_use] default fn le(&self, other: &Rhs) -> bool`
1.0.0
[src]

This method tests less than or equal to (for self and other) and is used by the <= operator. Read more

`#[must_use] default fn gt(&self, other: &Rhs) -> bool`
1.0.0
[src]

This method tests greater than (for self and other) and is used by the > operator. Read more

`#[must_use] default fn ge(&self, other: &Rhs) -> bool`
1.0.0
[src]

This method tests greater than or equal to (for self and other) and is used by the >= operator. Read more

`impl<T: Float> PartialEq<Tensor<T>> for Tensor<T>`[src]

`fn eq(&self, other: &Tensor<T>) -> bool`[src]

`#[must_use] default fn ne(&self, other: &Rhs) -> bool`
1.0.0
[src]

This method tests for !=.

`impl<T: Float> AsRef<Tensor<T>> for Tensor<T>`[src]

`fn as_ref(&self) -> &Tensor<T>`[src]

`impl<T: Float> Display for Tensor<T>`[src]

`fn fmt(&self, f: &mut Formatter) -> Result`[src]