Is there a library in js as powerful as numpy in python?

as mentioned in the question, is there any library in js that can perform matrix operations and operate fast like numpy?
if not, does it mean that js is not suitable for data operations such as mathematical statistics?

Mar.04,2021

as far as I know, js itself is not suitable for CPU-intensive computing, but there are node.js modules written in CPP that allow you to perform operations at the speed of CPP. In browsers, there are also libraries that use WebGL for acceleration, and there are packages to use WebGL in node.js. I searched at random and found numbers.js and math.js , but I don't know what the effect is.


there should be no js, this piece of js is almost blank, and the most existing matrix operations are in webgl, but basically, the matrices used in this part of the graph are all below the fourth order square moments. I think it may not be the
that you want. You can try to write your own efficiency. You can use webgl to communicate with gpu to accelerate matrix operations or SIMD can also

.

js can also do some simple data operations such as mathematical statistics, and there are also simple ai projects written in js
, but it can also be said that js is not suitable for data operations such as mathematical statistics. Py can be because there are a large number of libraries that support these libraries, most of the speed aspects are written in c languages, but on the contrary, js does not have these benefits. The biggest advantage of js is that the cross-platform environment is easy to configure. Unlike the problem that py is trying to face and solve,
nodejs may be able to write plug-ins in c/cpp after all, but everyone's enthusiasm is not there, so it's normal that libraries in this area are missing

.

numjs


as far as I know, the application of js in this area is still relatively few.

Menu