Standing on the shoulders of giants is a famous proverb. In software engineering, it reminds us of the importance of reusing resources that already exist in the public domain.
There is no doubt that TensorFlow.js is a powerful and practical machine learning library running in the JavaScript environment. Technically, we can build any kind of machine learning model by combining the operations implemented in TensorFlow.js. However, working with raw operations is not always easy, and not even the best practice. If you only desire to use pretty mature existing algorithms, implementing a model by yourself is actually not what you want to do. The libraries we are going to introduce already have implementations of the basic machine learning and deep learning algorithms so that you can try them immediately.
Another reason is for learning purposes. There are too...