TensorSpace.js

Present Tensor in Space

English | 中文

TensorSpace is a neural network 3D visualization framework built using TensorFlow.js, Three.js and Tween.js. TensorSpace provides Keras-like APIs to build deep learning layers, load pre-trained models, and generate a 3D visualization in the browser. From TensorSpace, it is intuitive to learn what the model structure is, how the model is trained and how the model predicts the results based on the intermediate information. After preprocessing the model, TensorSpace supports to visualize pre-trained model from TensorFlow, Keras and TensorFlow.js.

Fig. 1 - Interactive LeNet created by TensorSpace

Table of Content

Motivation
Getting Started
Awesome TensorSpace
Example
Documentation
Changelog
Contributors
Contact
License

Motivation

TensorSpace is a neural network 3D visualization framework designed for not only showing the basic model structure, but also presenting the processes of internal feature abstractions, intermediate data manipulations and final inference generations.

By applying TensorSpace API, it is more intuitive to visualize and understand any pre-trained models built by TensorFlow, Keras, TensorFlow.js, etc. TensorSpace introduces a way for front end developers to be involved in the deep learning ecosystem. As an open source library, TensorSpace team welcomes any further development on visualization applications.

Interactive -- Use Keras-like API to build interactive model in browsers.
Intuitive -- Visualize the information from intermediate inferences.
Integrative -- Support pre-trained models from TensorFlow, Keras, TensorFlow.js.

Getting Started

Fig. 2 - TensorSpace Workflow

1. Install TensorSpace Library

Basic Case

Option 1: From CDN

<script src="https://cdnjs.cloudflare.com/ajax/libs/three.js/97/three.min.js"></script>
<script src="https://cdnjs.cloudflare.com/ajax/libs/tween.js/17.2.0/Tween.min.js"></script>
<script src="https://cdnjs.cloudflare.com/ajax/libs/tensorflow/0.13.4/tf.min.js"></script>
<script src="https://tensorspace.org/assets/jslib/TrackballControls.js"></script>
<!-- Replace "VERSION"  with the version you want to use. -->
<script src="https://cdn.jsdelivr.net/npm/tensorspace@VERSION/build/tensorspace.min.js"></script>

Option 2: Download and include tensorspace.min.js in web page.

Get tensorspace.min.js from Github, NPM or TensorSpace official website

<script src="three.min.js"></script>
<script src="tween.min.js"></script>
<script src="tf.min.js"></script>
<script src="TrackballControls.js"></script>
<script src="tensorspace.min.js"></script>

Using TensorSpace in Progressive Framework

Step 1: Install TensorSpace
- Option 1: NPM
```
npm install tensorspace
```
- Option 2: Yarn
```
yarm add tensorspace
```
Step 2: Use TensorSpace

import * as TSP from 'tensorspace';

Checkout this Angular example for more information.

2. Preprocess Pre-trained Model

For presenting multiple intermediate outputs, we need to preprocess the pre-trained model.

Based on different training libraries, we provide different tutorials: TensorFlow model preprocessing, Keras model preprocessing and TensorFlow.js model preprocessing.

3. Create 3D TensorSpoace Model

If TensorSpace is installed successfully and the pre-trained deep learning model is preprocessed, let's create an interactive 3D TensorSpace model.

For convenience, feel free to use the resources from our HelloWorld directory.

We will use the preprocessed TensorSpace compatible LeNet model and sample input data ("5") as an example to illustrate this step. All source code can be found in helloworld.html.

First, we need to new a TensorSpace model instance:

let container = document.getElementById( "container" );
let model = new TSP.models.Sequential( container );

Next, based on the LeNet structure: Input + 2 X (Conv2D & Maxpooling) + 3 X (Dense), we build the structure of the model:

model.add( new TSP.layers.GreyscaleInput({ shape: [28, 28, 1] }) );
model.add( new TSP.layers.Padding2d({ padding: [2, 2] }) );
model.add( new TSP.layers.Conv2d({ kernelSize: 5, filters: 6, strides: 1 }) );
model.add( new TSP.layers.Pooling2d({ poolSize: [2, 2], strides: [2, 2] }) );
model.add( new TSP.layers.Conv2d({ kernelSize: 5, filters: 16, strides: 1 }) );
model.add( new TSP.layers.Pooling2d({ poolSize: [2, 2], strides: [2, 2] }) );
model.add( new TSP.layers.Dense({ units: 120 }) );
model.add( new TSP.layers.Dense({ units: 84 }) );
model.add( new TSP.layers.Output1d({
    units: 10,
    outputs: ["0", "1", "2", "3", "4", "5", "6", "7", "8", "9"]
}) );

Last, we should load our preprocessed TensorSpace compatible model and use init() method to create the TensorSpace model:

model.load({
    type: "tfjs",
    url: './lenetModel/mnist.json'
});
model.init(function(){
    console.log("Hello World from TensorSpace!");
});

We can get the following Fig. 2 model in the browser if everything looks good.

Fig. 3 - LeNet model without any input data

We provide a extracted file which is a handwritten "5" as the input of our model: (online demo)

model.init(function() {
    model.predict( image_5 );
});

We put the predict( image_5 ) method in the callback function of init() to ensure the prediction is after the initialization complete.

Click the CodePen logo to try it in CodePen:

Fig. 4 - LeNet model with input data "5"

Example

LeNet [ TensorFlow.js model ]

➡ Live Demo

Fig. 5 - Interactive LeNet created by TensorSpace

AlexNet [ TensorFlow model ]

➡ Live Demo

Fig. 6 - Interactive AlexNet created by TensorSpace

Yolov2-tiny [ TensorFlow model ]

➡ Live Demo

Fig. 7 - Interactive Yolov2-tiny created by TensorSpace

ResNet-50 [ Keras model ]

➡ Live Demo

Fig. 8 - Interactive ResNet-50 created by TensorSpace

Vgg16 [ Keras model ]

➡ Live Demo

Fig. 9 - Interactive Vgg16 created by TensorSpace

ACGAN [ Keras model ]

➡ Live Demo

Fig. 10 - Interactive ACGAN created by TensorSpace

MobileNetv1 [ Keras model ]

➡ Live Demo

Fig. 11 - Interactive MobileNetv1 created by TensorSpace

Inceptionv3 [ Keras model ]

➡ Live Demo

Fig. 12 - Interactive Inceptionv3 created by TensorSpace

LeNet Training Visualization [ TensorFlow.js dynamic model ]

Visualize the LeNet Training Process with TensorSpace.js and TensorFlow.js.

➡ Live Demo

Fig. 13 - LeNet Training 3D Visualization

View models locally

As some models above are extremely large, view them locally may be a good choice.

Step 1: clone TensorSpace Repo

git clone https://github.com/tensorspace-team/tensorspace.git

Step 2:

Open "html" file in examples folder in local web server.

Documentation

For a quick start, checkout Getting Start
To download/install, see Download
To learn more about the Basic Concepts
To process a deep learning model, checkout Model Preprocessing
To learn core components: Models, Layers and Merge Function
Checkout the official website TensorSpace.org for more about TensorSpace.

Contributors

Thanks goes to these wonderful people (emoji key):

_syt123450 💻 🎨 📖 💡	_{Chenhua Zhu} 💻 🎨 ✅ 💡	_{YaoXing Liu} 💻 🎨 ✅ 💡	_Qi(Nora) 💻 🎨	_{Dylan Schiemann} 📝	_BoTime 💻 📖 💡	_{Kamidi Preetham} 📖
_{Wade Penistone} 📖

Contact

If you have any issue or doubt, feel free to contact us by:

Email: tensorspaceteam@gmail.com
GitHub Issues: create issue
Slack: #questions
Gitter: #Lobby

License

Apache License 2.0

digitake/tensorspace

TensorSpace.js

Table of Content

Motivation

Getting Started

1. Install TensorSpace Library

2. Preprocess Pre-trained Model

3. Create 3D TensorSpoace Model

Example

View models locally

Documentation

Contributors

Contact

License