Image by : Christian Holmér. Modified by Opensource.com. CC BY-SA 4.0
The tremendous growth of the machine learning field has been driven by the availability of open source tools that allow developers to build applications easily. (For example, AndreyBu, who is from Germany and has more than five years of experience in machine learning, has been utilizing various open source frameworks to build captivating machine learning projects.)
TensorFlow.js is an open source library that allows you to run machine learning programs completely in the browser. It is the successor of Deeplearn.js, which is no longer supported. TensorFlow.js improves on the functionalities of Deeplearn.js and empowers you to make the most of the browser for a deeper machine learning experience. With the library, you can use versatile and intuitive APIs to define, train, and deploy models from scratch right in the browser. Furthermore, it automatically offers support for WebGL and Node.js.
If you have pre-existing trained models you want to import to the browser, TensorFlow.js will allow you do that. You can also retrain existing models without leaving the browser.
2. Machine learning tools
The machine learning tools library is a compilation of resourceful open source tools for supporting widespread machine learning functionalities in the browser. The tools provide support for several machine learning algorithms, including unsupervised learning, supervised learning, data processing, artificial neural networks (ANN), math, and regression.
Keras.js is another trending open source framework that allows you to run machine learning models in the browser. It offers GPU mode support using WebGL. If you have models in Node.js, you’ll run them only in CPU mode. Keras.js also offers support for models trained using any backend framework, such as the Microsoft Cognitive Toolkit (CNTK).
Some of the Keras models that can be deployed on the client-side browser include Inception v3 (trained on ImageNet), 50-layer Residual Network (trained on ImageNet), and Convolutional variational auto-encoder (trained on MNIST).
The library comes with comprehensive and advanced mathematical and statistical functions to assist you in building high-performing machine learning models. You can also use its expansive utilities for building applications and other libraries. Furthermore, if you want a framework for data visualization and exploratory data analysis, you’ll find STDLib worthwhile.
Do you know of another open source library that offers in-browser machine learning capabilities? Please let us know in the comment section below.