Struct Tensor 

pub struct Tensor {
    pub value: Array1<f64>,
    pub grad: Option<Array1<f64>>,
}

Represents a basic implementation of a tensor.

Fields

value: Array1<f64>

grad: Option<Array1<f64>>

Implementations

impl Tensor

pub fn new(value: Array1<f64>) -> Tensor

Creates a new Tensor with the given value.

Arguments

• value - The array representing the value of the tensor.

Returns

(Tensor): A new Tensor instance.

Examples

use ndarray::{array, Array1};
use tinygrad::Tensor;

let value = array![1.0, 2.0, 3.0];
let tensor = Tensor::new(value);

Trait Implementations

impl Clone for Tensor

fn clone(&self) -> Tensor

Returns a duplicate 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 TensorTrait for Tensor

fn forward(&self, _ctx: &mut Context, _inputs: Vec<ArrayView1<'_, f64>>) -> f64

Implements the forward pass for the Tensor.

Arguments

• _ctx - A mutable reference to the computation context (not used in this example).
• _inputs - A vector of ArrayView1 representing the input values (not used in this example).

Returns

(f64): The result of the forward pass (not implemented in this example, returns 0.0).

fn backward(&self, _ctx: &mut Context, _grad_output: ArrayView1<'_, f64>)

Implements the backward pass for the Tensor.

Arguments

• _ctx - A mutable reference to the computation context (not used in this example).
• _grad_output - An ArrayView1 representing the gradient of the output.

Examples

use ndarray::{array, Array1};
use tinygrad::{Tensor, Context, TensorTrait};

let mut ctx = Context::new();
let tensor = Tensor::new(array![1.0, 2.0, 3.0]);
tensor.backward(&mut ctx, array![1.0, 1.0, 1.0].view());

fn get_value(&self) -> ArrayView1<'_, f64>

Returns the value of the Tensor.

Returns

(ArrayView1<f64>): The view of the array representing the value of the tensor.

Examples

use ndarray::{array, Array1};
use tinygrad::{Tensor, TensorTrait};

let tensor = Tensor::new(array![1.0, 2.0, 3.0]);
let value = tensor.get_value();

fn get_grad(&self) -> Option<Array1<f64>>

Returns the gradient of the Tensor if available.

Returns

(Option<Gradient>): The optional gradient of the tensor.

Examples

use ndarray::{array, Array1};
use tinygrad::{Tensor, TensorTrait};

let tensor = Tensor::new(array![1.0, 2.0, 3.0]);
let grad = tensor.get_grad();