3.3 KiB
3.3 KiB
SVID算法文章代码验证步骤(SVIDAlgorithm article code verification steps)
Step 01 环境准备 Environment Preparation
- 需要安装Matlab
- Need to install Matlab
Step 02 下载代码 Download the code
- 将
Code_for_Paper
和Original_Code
两个文件夹拉取到本地。 - Pull the two folders
Code_for_Paper
andOriginal_Code
to your local computer.
Step 03 下载数据 Download data
- 由于Github对上传文件大小有限制,我将会用到的数据存在云端,可以通过链接进行下载,里面应该含有
Test1SimandExp.mat
、Test2SimandExp_1.mat
、Test2SimandExp_2.mat
和Test3SimandExp.mat
四个文件。检查无误后将其分别复制到下载好的Code_for_Paper
和Original_Code
文件夹中。 - Since Github has a limit on the size of uploaded files, the data I will use is stored in the cloud and can be downloaded through link, which should contain four files:
Test1SimandExp.mat
,Test2SimandExp_1.mat
,Test2SimandExp_2.mat
, andTest3SimandExp.mat
. After checking that they are correct, copy them to the downloadedCode_for_Paper
andOriginal_Code
folders respectively.
Step 04 运行论文所使用代码 Run the code used in the paper
- 在Matlab环境中运行
Code_for_Paper
文件夹中Test1Plot.m
即可获得数据修改版本的Fig.A1-A3,运行Test2Plot.m`即可获得数据修改版本的Fig.A4-A5,运行
Test3Plot.m`即可获得数据修改版本的Fig.A6-A7 - Run
Test1Plot.m
in theCode_for_Paper
folder in the Matlab environment to get the modified version of Fig.A1-A3, runTest2Plot.m` to get the modified version of Fig.A4-A5, run
Test3Plot.m` to get the modified version of Fig.A6-A7,
Step 05 运行原始代码 Run the original code
- 在Matlab环境中运行
Original_Code
文件夹中Test1Plot.m
即可获得原本数据版本的Fig.A1-A3,运行Test2Plot.m`即可获得原本数据版本的Fig.A4-A5,运行
Test3Plot.m`即可获得原本数据版本的Fig.A6-A7, - Run
Test1Plot.m
in theOriginal_Code
folder in the Matlab environment to get the original data version of Fig.A1-A3, runTest2Plot.m` to get the original data version of Fig.A4-A5, run
Test3Plot.m` to get the original data version of Fig.A6-A7,
Step 06 对比分析 Comparative analysis
- 比较两者结果差异,验证“学术不端行为”的有效性。
- Compare the differences between the two results to verify the effectiveness of "academic misconduct".
Step 07 一点技巧 A little trick
- 细心的同学可能发现得到的图和论文中的图在尺寸和线条颜色上有不同,那是因为我担心被导师发现,先在matlab中弄好数据放缩后,将数据导入到其他画图软件,例如Origin里面重新绘图,这样导师就不可能发现了。
- Careful people may find that the obtained figures are different from those in the paper in size and line color. That’s because I was worried that my supervisor would find out, so I first scaled the data in Matlab, and then imported the data into other drawing software, such as Origin, and re-drew the figure. This way, my supervisor would not be able to find out.