preface

I found GoPlus on Github. It is said that GoPlus is used in data science, so this should be a tagged blog. I have no time to learn more about this content recently, so I write that I will learn about it when I have time in the future.

Introduction to the

We do data analysis and data science, mostly in Python, Julia, or R, but it was interesting to see that GO could do data science, so I took a look.Let’s start with the Github link:github.com/goplus/gop If you look at readme, goPlus makes go code more suitable for data science writing, is compatible with Go, and even simplifies the amount of code. It should be easy for people who are used to writing Python to get started, eliminating the need for type declarations when variables are declared in go itself (it’s still a static language). Of course, there are corresponding tutorials, but I do not know whether the time is not enough or the current development is not complete, see the content of the tutorial only feel too simple, basic is to give some examples to take over, more need to give their own code examples to understand their own. The sudden 3:3.4 in the list below, for example, can be confusing at first.Of course, like most languages, we can still run our code online without a local environment. Here is the website of the online platform:Qiniu. Making. IO/goplus – play… In addition to GoPlus, Go has a numpy-like librarynumgo Github.com/Kunde21/num… Ndarray was implemented using Go, which is not too nice to like using Numpy.

At the end

Although I just looked at it briefly, I still had a lot of impressions. First of all, we are computer students, and our knowledge is changing so fast that maybe Python is hot today, but maybe in five years there will be a simpler, more attractive language that will lead the way. There’s so much to learn, so much more to learn, and time management. Getting back to today’s topic, whether it’s GoPlus or NumGO, I believe the contributors decided to develop a data science language based on GO out of a combination of their love of the language and the performance, speed, and difficulty of coding it. It’s true that Go has much better performance than Python and is a little more difficult to code, so using Go for data science might be a good choice, but there are still plenty of drawbacks. The course is not complete, the whole ecology is not complete. People like Python because there are too many libraries to help us save a lot of coding time. For machine learning and deep learning, we may not know much about the principle, as long as we can call the API of the relevant framework. On the contrary, go’s ecology in the field of data science is not as good as Python (mention the GORo library of Go here. Very similar to Keras, but also packaged into a high-level API), so there are still few people who know about it, and even fewer people choose to use it. I hope GO can improve the ecology. Personally, I like GO very much, and I want to make some contributions to something I am interested in when I have time.