Shulou(Shulou.com)06/02 Report--

This article mainly introduces "what are the Javascript machine learning libraries worth studying". In the daily operation, I believe many people have doubts about the Javascript machine learning libraries worth studying. The editor consulted all kinds of materials and sorted out simple and easy-to-use operation methods. I hope it will be helpful to answer the questions of "what are the Javascript machine learning libraries worth studying?" Next, please follow the editor to study!

1. Brain.js

Brain.js is the Javascript library of neural networks, replacing the (now deprecated) "brain" library, which can be used with Node.js or in browsers (note computing), and provides different types of networks for different tasks.

Github address: https://github.com/BrainJS/brain.js

This is a demonstration to train the network to recognize color contrast.

2. Synaptic

Synaptic is a Javascript neural network library for node.js and browsers that enables you to train first-and even second-level neural network architectures. The project includes a number of built-in structures, such as multi-layer perceptrons, multi-layer long-and short-term memory networks, liquid state machines and trainers who can train real networks.

Github address: https://github.com/cazala/synaptic

3. Neataptic

The library provides rapid neural evolution and back propagation for browsers and Node.js, including some built-in networks, including perceptrons, LSTM,GRU,Nark, etc. This is a novice tutorial for simple training.

Github address: https://github.com/wagenaartje/neataptic

4. ConvNetJS

Developed by a doctor from Stanford University, this popular library has not been maintained for the past four years, but it is definitely one of the most interesting items on the list. It is the Javascript implementation of neural network, supporting general modules, classification, regression, experimental reinforcement learning modules, and even training convolution networks for image processing.

Github address: https://github.com/karpathy/convnetjs

It currently supports:

Common neural network modules (fully connected layer, nonlinear)

Classification (SVM / Softmax) and regression (L2) cost function

Ability to specify and train convolution networks for image processing

Experimental reinforcement Learning Module based on Deep Q Learning

5. WebDNN

This Japanese-made JavaScript library is built to quickly run deep neural network pre-training models on browsers. Because executing DNN on browsers consumes a lot of computing resources, the framework optimizes the DNN model to compress model data and speed up execution through JavaScript API, such as WebAssembly and WebGPU.

An example of using WebDNN to do neural network style transfer:

6. Deeplearnjs

This popular library allows you to train neural networks in browsers or run pre-trained models in inference mode, even claiming that it can be used as a NumPy for the network. With the easily available API, the library can be used in practical and useful applications and can be actively maintained.

Github address: https://github.com/tensorflow/tfjs-core

As part of the TensorFlow.js ecosystem, it is now hosted on @ tensorflow/tfjs-core and is the core API of TensorFlow.js, which provides low-level, hardware-accelerated linear algebraic operations and urgent API for automatic differentiation.

7. Tensorflow Deep Playground

Tensorflow Deep Playground is an interactive visualization of neural network, written in d3.js and TypeScript. Although this project basically contains a very basic tensor flow playground, it can be used for different means or for very impressive educational functions for different purposes.

Github address: https://github.com/tensorflow/playground

8. Compromise

This very popular library provides "moderate natural language processing in javascript". This is so basic and straightforward that it can even be compiled into a small file. For some reason, its moderately "good enough" approach makes it the main candidate for almost any application that requires basic NLP.

Github address: https://github.com/spencermountain/compromise

9. Neuro.js

This beautiful project is a browser for deep learning and reinforcement learning Javascript library framework. To implement a machine learning framework based on full stack neural network and extended reinforcement learning support, some people think that this project is the successor of convnetjs.

Github address: https://github.com/janhuenermann/neurojs

10. Ml.js

A set of repositories that provide machine learning tools for Javascript developed by mljs, including supervised and unsupervised learning, artificial neural networks, regression algorithms and statistics, mathematics, and other support libraries. This is a short exercise.

Github address: https://github.com/mljs

11. Mind

A flexible neural network library for Node.js and browsers, basically learning how to make predictions, using matrix implementations to process training data and enabling configurable network topologies. You can also plug and play "ideas" that you have learned, which is very useful for your application.

Github address: https://github.com/stevenmiller888/mind

At this point, the study of "what are the Javascript machine learning libraries worth studying" is over. I hope to be able to solve your doubts. The collocation of theory and practice can better help you learn, go and try it! If you want to continue to learn more related knowledge, please continue to follow the website, the editor will continue to work hard to bring you more practical articles!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.