Tensorflow.js tf.regularizers.l1() Function
Last Updated :
27 Jul, 2021
Tensorflow.js is an open-source library that is developed by Google for running machine learning models as well as deep learning neural networks in the browser or node environment.
The .regularizers.l1() function is used for L1 regularization. Moreover, it appends a name to the loss in order to rebuke enormous weights: loss += sum(l1 * abs(x)).
Syntax:
tf.regularizers.l1(config?)
Parameters:
- config: It is an object which is optional. And under it comes l1.
- l1: It is the stated L1 regularization rate, whose by default value is 0.01. It is of type number.
Return Value: It returns Regularizer.
Example 1: In this example, we are going to see the standalone use of l1 Regularizer applied to the kernel weights matrix.
Javascript
const tf = require( "@tensorflow/tfjs" );
const model = tf.sequential();
model.add(tf.layers.dense({
units: 37, batchInputShape:[ null , 40],
kernelRegularizer:tf.regularizers.l1()
}));
model.summary();
|
Output:
_________________________________________________________________
Layer (type) Output shape Param #
=================================================================
dense_Dense52 (Dense) [null,37] 1517
=================================================================
Total params: 1517
Trainable params: 1517
Non-trainable params: 0
_________________________________________________________________
Example 2: In this example, we are going to see the standalone use of l1 Regularizer applied to the bias vector.
Javascript
const tf = require( "@tensorflow/tfjs" );
const model = tf.sequential();
model.add(tf.layers.dense({
units: 2, batchInputShape:[ null , 13],
biasRegularizer:tf.regularizers.l1()
}));
model.summary();
|
Output:
_________________________________________________________________
Layer (type) Output shape Param #
=================================================================
dense_Dense54 (Dense) [null,2] 28
=================================================================
Total params: 28
Trainable params: 28
Non-trainable params: 0
_________________________________________________________________
Reference: https://js.tensorflow.org/api/latest/#regularizers.l1
Share your thoughts in the comments
Please Login to comment...