Tensorflow.js tf.Sequential class.predict() Method
Last Updated :
04 Jun, 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 .predict() method is used to produce the output expectations considering the stated input instances. Moreover, the calculation is performed here in groups.
Syntax:
predict(x, args?)
Parameters:
- x: It is the stated input data, like a tensor, or else an array of tf.Tensors in case the model has numerous inputs. It can be type tf.Tensor, or tf.Tensor[].
- args: It is an optional parameter that is of type object.
- batchSize: It is the stated batch size in integer form and is optional parameter. Moreover, in case its not specified, then the default value is 32.
- verbose: It is the stated verbosity mode which is by default false. It is of type Boolean and is optional.
Return Value: It returns tf.Tensor or tf.Tensor[].
Example 1:
Javascript
import * as tf from "@tensorflow/tfjs"
const modl = tf.sequential({
layers: [tf.layers.dense({units: 2, inputShape: [40]})]
});
modl.predict(tf.truncatedNormal([7, 40])).print();
|
Output:
Tensor
[[0.1556173 , 1.2365075 ],
[-1.7945877, 2.3424799 ],
[0.3632407 , -0.1731701],
[0.195157 , -0.7823027],
[0.4565429 , 2.512109 ],
[-1.2392806, 0.1868197 ],
[0.6895663 , -0.2006246]]
Example 2:
Javascript
import * as tf from "@tensorflow/tfjs"
tf.sequential({
layers: [tf.layers.dense({units: 3, inputShape: [20]})]
}).predict(tf.randomNormal([8, 20])).print();
|
Output:
Tensor
[[-1.1149288, 0.8968778 , -0.7492741],
[1.3654473 , -0.471923 , 1.3632329 ],
[0.5550661 , 0.6949158 , 1.9761562 ],
[-0.2109454, -0.3558912, 0.243051 ],
[-1.2827762, 0.5370077 , 0.1645843 ],
[0.1542411 , -1.359634 , -0.1656512],
[-0.4721956, 0.3904444 , 0.7398967 ],
[-0.2076109, -3.0447464, 1.3338548 ]]
Reference: https://js.tensorflow.org/api/latest/#tf.Sequential.predict
Share your thoughts in the comments
Please Login to comment...