How the most famous DeFi project processes data in the repository
The main purpose of this article is to analyze the repository of the most popular DeFi projects through data science, based on data analysis to help those who want to collaborate or contribute to creating their own DeFi project.
This article only shows the analysis of the acquired data. The analyzed items are shown on the following DeFi project map.
Mapping out Ethereum’s DeFi
Data set description
In order to evaluate the project, this article fetches data through the Github API and compiles all the repositories that make up each project, excluding branches of other projects. A total of 1588 GitHub code repositories were analyzed.
Sometimes it seems easy to create an item like in a picture. We only need to spend a small amount of work to see some data that make us have a more realistic impression of the meaning of the project.
By looking at the repositories that make up a project, we find that there are an average of 14 repositories per project and an average of 21.5 contributors per project.
If we look in more detail, we will find which projects have more repositories, and even some of the most famous projects have more than 100 repositories.
In these repositories, people participate in the development of repositories. Although people can collaborate multiple times in the same repository, if we look at how many unique contributors each project has and add them, we see that in some cases these are quite large projects.
In addition, we found that reaching their current position did not happen overnight. If we see the creation time of the first repository for these projects, it is likely that it will require a long growth period like all large projects.
The following table shows the name of the project and the creation date of its first repository, in ascending order.
It is generally believed that to create a blockchain project, a large amount of code must be developed on the blockchain, and most of the work will focus on writing stable and reliable contracts.
If we look at the major languages in these project repositories, we see:
To analyze this in more detail, we will get all the languages that appear in the repository. Because GitHub provides a certain percentage for each repository and assigns the dominant language to the repository with the most projects.
For example, in this case, we would say that the Python language is used, although there are other languages with sufficient weight:
If we get the weights in each repository and add them up, we see something like the above:
Popularity of the project
To analyze how popular these projects are among developers, we will look for how many star projects are in the repository, how many star projects have been created branches, and how many people have subscribed to them in order to notify people of changes in them.
In open source projects, you can collaborate by reporting or helping with code issues. Once this issue is resolved, it will "merge" within the code and close the error. We can check which projects have more open issues:
A good way to collaborate with these projects is to help solve problems. First, it's best to look for projects that are classified as "good for the first problem" so that you can collaborate and learn from the project, which reduces complexity or does not require in-depth knowledge of the project. .
https://gist.github.com/ialberquilla/05e1511134e2486f96bfed8cf0470667, I have put this kind of problems and the projects they belong to in a list, I hope to help you.