TensorSpace.js
Getting Start
Basic Concepts
TensorSpace Converter
Model Preprocessing
Models
Layers
Merge Function
Conv2dTranspose
Transpose convolution (also known as deconvolution) is a partial same functional with upsampling. GAN model use it to construct the generative network.
Constructor
Based on whether TensorSpace Model load a pre-trained model before initialization, configure Layer in different ways. Checkout Layer Configuration documentation for more information about the basic configuration rules.
〔Case 1〕If TensorSpace Model has loaded a pre-trained model before initialization, there is no need to configure network model related parameters.
TSP.layers.Conv2dTranspose();
〔Case 2〕If there is no pre-trained model before initialization, it is required to configure network model related parameters.
[Method 1] Use filters, kernelSize and strides
TSP.layers.Conv2dTranspose( { filters : Int, kernelSize: Int, strides: Int } );
[Method 2] Use shape
TSP.layers.Conv2dTranspose( { shape : [ Int, Int, Int ] } );
Fig. 1 - Conv2dTranspose layer close and open
Arguments

Name

Tag

Type

Instruction

Usage Notes and Examples

name

String Name of the layer For example, name: "layerName"
In Sequential Model: Highly recommend to add a name attribute to make it easier to get Layer object from model.
In Functional Model: It is required to configure name attribute for TensorSpace Layer, and the name should be the same as the name of corresponding Layer in pre-trained model.

filters

Int Amount of filters,
Network Model Related
filters: 16

padding

String Padding mode,
Network Model Related

valid[default]: Without padding. It drops the right-most columns (or bottom-most rows).

same: With padding. Output size is the same as input size.

kernelSize

Int | Int[] The dimension of the convolution window,
Network Model Related
The 2d convolutional window. For example,
Square kernel, configre kernelSize to be 3
Rectangle kernel, configure kernelSize to [1, 7].

strides

Int | Int[] The strides of the convolution,
Network Model Related
Strides in both dimensions, [default] Automatic inference strides: [1, 1]. For example,
Strides in width and height is same, configure strides to be 2
Strides in width and height is different, configure strides to be [1, 2]

shape

Int[] Output shape,
Network Model Related
For example, shape: [ 28, 28, 6 ]
means the output is 3-dimensional, 6 feature maps and each one is 28 by 28. Data format is channel last.

color

Color Format Color of layer Conv2dTranspose's default color is #FF5722

closeButton

Dict Close button appearance control dict. More about close button

display: Bool. true[default] Show button, false Hide button

ratio: Int. Times to close button's normal size, default is 1, for example, set ratio to be 2, close button will become twice the normal size

initStatus

String Layer initial status. Open or Close. More about Layer initial Status close[default]: Closed at beginning,
open: Open at beginning

animeTime

Int The speed of open and close animation For example, animeTime: 2000
means the animation time will last 2 seconds.
Note: Configure animeTime in a specific layer will override model's animeTime configuration.
Properties
.inputShape : Int[]
filter_center_focusThe shape of input tensor, for example inputShape = [ 28, 28, 3 ] represents 3 feature maps and each one is 28 by 28.
filter_center_focusAfter model.init() data is available, otherwise is undefined.
.outputShape : Int[]
filter_center_focusThe shape of output tensor is 3-dimensional. 3️⃣
filter_center_focusdataFormat is channel last. for example outputShape = [ 32, 32, 4 ] represents the output through this layer has 4 feature maps and each one is 32 by 32
filter_center_focusAfter model.init() data is available, otherwise is undefined.
.neuralValue : Float[]
filter_center_focusThe intermediate raw data after this layer.
filter_center_focusAfter load and model.predict() data is available, otherwise is undefined.
.name : String
filter_center_focusThe customized name of this layer.
filter_center_focusOnce created, you can get it.
.layerType : String
filter_center_focusType of this layer, return a constant: string Conv2dTranspose.
filter_center_focusOnce created, you can get it.
Methods
filter_center_focusLink this layer to previous layer.
filter_center_focusThis method can be use to construct topology in Functional Model.
filter_center_focus See Construct Topology for more details.
.openLayer() : void
filter_center_focus Open Layer, if layer is already in "open" status, the layer will keep open.
filter_center_focus See Layer Status for more details.
filter_center_focus Close Layer, if layer is already in "close" status, the layer will keep close.
filter_center_focus See Layer Status for more details.
Examples
filter_center_focus If TensorSpace Model load a pre-trained model before initialization, there is no need to configure network model related parameters.
let transposeLayer = new TSP.layers.Conv2dTranspose( {

    // Recommend Configuration. Required for TensorSpace Functional Model.
    name: "conv2dTranspose1",
    // Optional Configuration.
    animeTime: 4000,
    initStatus: "open"

} );
filter_center_focus If there is no pre-trained model before initialization, it is required to configure network model related parameters.
let transposeLayer = new TSP.layers.Conv2dTranspose( {

    // Required network model related Configuration.
    kernelSize: 5,
    filters: 6,
    strides: 1,
    // Recommend Configuration. Required for TensorSpace Functional Model.
    name: "conv2dTranspose2",
    // Optional Configuration.
    animeTime: 4000,
    initStatus: "open"

} );
Use Case
When you add de-convolution layer with Keras | TensorFlow | tfjs in your model the corresponding API is Conv2dTranspose in TensorSpace.
Framework Documentation
Keras keras.layers.Conv2DTranspose(filters, kernel_size, strides=(1, 1))
TensorFlow tf.nn.conv2d_transpose(value, filter, output_shape, strides)
TensorFlow.js tf.layers.conv2dTranspose (config)
Source Code
tensorspace/src/layer/intermediate/Conv2dTranspose.js