Blog posts

2022

Writing Release Notes for Your Software: How to Get it Right

7 minute read

Published:

Release note is important. However, there is a lack of tutorials or widely acknowledged standards about how to produce a release note. Without “the right way,” release notes may cause all kinds of issues. In this article, we will provide an FAQ-style introduction on how to produce the “right” release note for your users, based on our recent research on ~1000 real release note issues in GitHub project. This is still a preliminary draft, so if you have any suggestions or critiques, feel free to comment below!

2021

开源软件量化分析研究入门

7 minute read

Published:

(出自我为我们实验室撰写的内部资料,是在周明辉老师写的某个早期版本上扩写而成,并增加了一点我自己的思考,因感觉非常有用,故在此留存)

2019

Mining GitHub Repository Information using the Official REST API

2 minute read

Published:

GitHub provides a (not very convinent and well documented) HTTP API for requesting information from GitHub. We can use https://api.github.com/search/repositories for requesting repository information in JSON format. You can apply various search conditions and sort them if necessary. For example, if you want to collect 1000 most starred repositories whose language is Java, you can use the following request.

基于五阶段流水线的RISC-V CPU模拟器实现

16 minute read

Published:

RISC-V是源自Berkeley的开源体系结构和指令集标准。这个模拟器实现的是RISC-V Specification 2.2中所规定RV64I指令集,基于标准的五阶段流水线,并且实现了分支预测模块和虚拟内存模拟。实现一个完整的CPU模拟器可以很好地锻炼系统编程能力,并且加深对体系结构有关知识的理解。在开始实现前,应当阅读并深入理解Computer Systems: A Programmer’s Perspective中的第四章,或者Computer Organizaton and Design: Hardware/Software Interface中的有关章节。

2018

Building Event System in Unity3D

3 minute read

Published:

When I was developing a simple 3D game using Unity 3D, I found it non-trivial to build an event system that could handle dynamic game events efficiently and elegantly.

隐马尔可夫模型(Hidden Markov Model)

1 minute read

Published:

隐马尔可夫模型(Hidden Markov Model, HMM)是一个重要的机器学习模型。直观地说,它可以解决一类这样的问题:有某样事物存在一定的状态,但我们无法得知某个时刻(或位置)它所处在的状态,但是我们有一个参照事物,我们知道这个参照事物在某个时刻(或位置)的状态并认为参照事物的状态和原事物的状态存在联系,那么我们可以使用机器学习来推测原事物最有可能在一个时刻(或位置)处在什么样的状态。也就是说,这是一个基于概率统计的模型。

真实感海洋的绘制(三):水面的渲染

less than 1 minute read

Published:

根据上述两篇博客,我们已经得到了真实感较高的水面波形和法向量。为了节省所需要的顶点数目,需要将高度场和法线制作成贴图传入着色器,以便从较少的顶点就能渲染很大面积的海面。之后的任务就是对这些波形进行真实感的渲染。

真实感海洋的绘制(二):使用快速傅里叶变换加速波形计算

less than 1 minute read

Published:

其实上一篇博文所写的$H(\vec{x},t)​$,就是二维傅里叶变换的求和式,之前的暴力计算法属于二维的离散傅里叶变换(Discrete Fourier Transform, DFT),利用二维的快速傅里叶变换(Fast Fourier Transform, FFT)可以将复杂度从$O(n^4)​$降低到$O(n^2\log{n})​$。

真实感海洋的绘制(一):基于统计学模型的水面模拟方法

less than 1 minute read

Published:

学习了基本的OpenGL和图形学知识后,第一个想做的事情就是画水(笑),因为对我而言各种游戏里面往往最令人印象深刻的就是那波光粼粼、使人心旷神怡的海面了~当然,海面的模拟并不是一件简单的事情TAT…因此决定对于其中较为一些复杂的内容整理出来发在博客上,供以后参考。

递推方程的求解

5 minute read

Published:

递推方程的求解是分析递归算法的一个重要步骤。