Introduction: 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 .conv3dTranspose() function is used to determine the transposed 3D convolution of a volume. It is also known as a deconvolution.
Syntax:
tf.conv3dTranspose(x, filter, outputShape, strides, pad)
Parameters:
- x: The stated input image which is either of rank 5 or else rank 4 and of shape: [batch, depth, height, width, inDepth]. Moreover, in case the rank is 4, then the batch of size 1 is presumed. It can be of type tf.Tensor4D, tf.Tensor5D, TypedArray, or Array.
- filter: The stated filter tensor of rank 4 and shape: [depth, filterHeight, filterWidth, outDepth, inDepth]. Where, inDepth must match inDepth in input tensor. It can be of type tf.Tensor5D, TypedArray, or Array.
- outputShape: The stated output shape which is of rank 5 or else rank 4 and shape [batch, depth, height, width, outDepth]. In case, the rank is 3, then the batch of 1 is presumed. It can be of type [number, number, number, number, number] or [number, number, number, number].
- strides: The stated strides of the original convolution of shape: [strideDepth, strideHeight, strideWidth]. It can be of type [number, number, number], or number.
- pad: The stated type of algorithm for padding which is useful in the non transpose form of the op. It can be of type valid, or same.
Return Value: It returns tf.Tensor4D or tf.Tensor5D.
Example 1:
Javascript
// Importing the tensorflow.js library import * as tf from "@tensorflow/tfjs" // Defining input tensor const x = tf.tensor5d([1, 2, 2, 3], [2, 2, 1, 1, 1]); // Defining filter tensor const y = tf.tensor5d([3, 3, 3, 2], [1, 2, 2, 1, 1]); // Calling conv3dTranspose() method const result = tf.conv3dTranspose(x, y, [1, 1, 2, 1, 1], 2, 'same'); // Printing output result.print(); |
Output:
Tensor
[[[ [[3],],
[[3],]]]]
Example 2:
Javascript
// Importing the tensorflow.js library import * as tf from "@tensorflow/tfjs" // Calling conv3dTranspose() method tf.conv3dTranspose(tf.tensor5d( [1.1, 2.1, 2.2, 3.6], [2, 2, 1, 1, 1]), tf.tensor5d([3.6, 3.1, 3.2, 2.0], [1, 2, 2, 1, 1]), [1, 1, 2, 1, 1], 7, 'valid').print(); |
Output:
Tensor
[[[ [[0],],
[[0],]]]]
Reference: https://js.tensorflow.org/api/latest/#conv3dTranspose
Whether you’re preparing for your first job interview or aiming to upskill in this ever-evolving tech landscape, neveropen Courses are your key to success. We provide top-quality content at affordable prices, all geared towards accelerating your growth in a time-bound manner. Join the millions we’ve already empowered, and we’re here to do the same for you. Don’t miss out – check it out now!
