Tensorflow.js tf.grad()函数
Tensorflow.js是谷歌开发的一个开源库,用于在浏览器或节点环境中运行机器学习模型和深度学习神经网络。
tf.grad()函数用于返回指定函数f(x) 相对于 x 的梯度。
句法:
tf.grad (f)
参数:此函数接受一个参数,如下所示:
- f:正在计算梯度的指定函数f(x)。
返回值:返回指定函数f(x) 相对于 x 的梯度。
示例 1:
Javascript
// Importing the tensorflow.js library
import * as tf from "@tensorflow/tfjs"
// Initializing a function f(x) = x^2
const f = x => x.square();
// Calling the .grad() function which
// calculates f'(x) and gives value as 2x
const g = tf.grad(f);
// Initializing a tensor of values for
// which gradient will be returned
const x = tf.tensor1d([1, 2, 3]);
// Getting the values of gradient of the
// function f(x) for the above specified
// tensor
g(x).print();
Javascript
// Importing the tensorflow.js library
import * as tf from "@tensorflow/tfjs"
// Using the function f(x) = x^3 as the
// parameter for the .grad() function which
// calculates f'(x) and gives value as 3x^2
const g = tf.grad(x => x.pow(tf.scalar(3, 'int32')));
// Initializing a tensor of values for
// which gradient will be returned
const x = tf.tensor1d([0, 1, 2]);
// Getting the values of gradient of the
// function f(x) for the above specified
// tensor
g(x).print();
输出:
Tensor
[2, 4, 6]
示例 2:
Javascript
// Importing the tensorflow.js library
import * as tf from "@tensorflow/tfjs"
// Using the function f(x) = x^3 as the
// parameter for the .grad() function which
// calculates f'(x) and gives value as 3x^2
const g = tf.grad(x => x.pow(tf.scalar(3, 'int32')));
// Initializing a tensor of values for
// which gradient will be returned
const x = tf.tensor1d([0, 1, 2]);
// Getting the values of gradient of the
// function f(x) for the above specified
// tensor
g(x).print();
输出:
Tensor
[0, 3, 12]
参考: https://js.tensorflow.org/api/latest/#grad