📜  Tensorflow.js tf.layers getConfig() 方法

📅  最后修改于: 2022-05-13 01:56:48.791000             🧑  作者: Mango

Tensorflow.js tf.layers getConfig() 方法

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.layers getConfig()函数用于获取层的配置。

句法:

getConfig()

参数:此函数不接受任何参数。

返回值:该函数返回层的配置。

例1:我们将在这个例子中得到最小层的配置。

Javascript
const tf = require("@tensorflow/tfjs")
 
// Creating a minLayer
const minLayer = tf.layers.minimum();
 
// Getting the configuration of the layer
const config = minLayer.getConfig();
console.log(config)


Javascript
const tf = require("@tensorflow/tfjs")
 
// Creating a minLayer
const denseLayer = tf.layers.dense({units: 1});
 
// Getting the configuration of the layer
const config = denseLayer.getConfig();
console.log(config)


Javascript
const tf = require("@tensorflow/tfjs")
 
// Creating a minLayer
const activationLayer =
    tf.layers.activation({activation: 'relu6'});
 
// Getting the configuration of the layer
const config = activationLayer.getConfig();
console.log(config)


输出:

{ name: 'minimum_Minimum1', trainable: true }

示例 2:我们将在此示例中获取密集层的配置。

Javascript

const tf = require("@tensorflow/tfjs")
 
// Creating a minLayer
const denseLayer = tf.layers.dense({units: 1});
 
// Getting the configuration of the layer
const config = denseLayer.getConfig();
console.log(config)

输出:

{ units: 1,
  activation: 'linear',
  useBias: true,
  kernelInitializer:
   { className: 'VarianceScaling',
     config:
      { scale: 1, mode: 'fanAvg', 
      distribution: 'normal', seed: null } },
  biasInitializer: { className: 'Zeros', config: {} },
  kernelRegularizer: null,
  biasRegularizer: null,
  activityRegularizer: null,
  kernelConstraint: null,
  biasConstraint: null,
  name: 'dense_Dense1',
  trainable: true }

示例 3:我们将在此示例中获取应用层的配置。

Javascript

const tf = require("@tensorflow/tfjs")
 
// Creating a minLayer
const activationLayer =
    tf.layers.activation({activation: 'relu6'});
 
// Getting the configuration of the layer
const config = activationLayer.getConfig();
console.log(config)

输出:

{ activation: 'relu6',
  name: 'activation_Activation1',
  trainable: true }

参考: https://js.tensorflow.org/api/latest/#tf.layers.Layer.getConfig