This repo is used to store other algorithm repo may be used in EGG project.

https://github.com/spolsley/common-spatial-patterns

https://github.com/erlendd/covariate-shift-adaption/blob/master/Supervised%20classification%20by%20covariate%20shift%20adaption.ipynb

https://www.arangodb.com/2020/11/arangoml-part-4-detecting-covariate-shift-in-datasets/

https://github.com/mne-tools/mne-python

EEG数据处理步骤：https://zhuanlan.zhihu.com/p/142068015

图神经网络:https://medium.com/@BorisAKnyazev/tutorial-on-graph-neural-networks-for-computer-vision-and-beyond-part-1-3d9fada3b80d

ROC曲线，AOC值：https://zhuanlan.zhihu.com/p/31256633

安装以及教程https://www.bilibili.com/video/BV1hE411t7RN?from=search&seid=15984300105231899247 注意：anaconda一定要安装64位的

https://github.com/caosen469/EGG_related/tree/main/pytorch_example

vgg_false = torchvision.models.vgg16(pretrained=False) # 只是加载vgg16的结构 vgg_true = torchvision.models.vgg16(pretrained=True) # 加载了结构，也加载了训练的参数

修改模型： 首先打印出网络结构，知道某层，或者某好几层构成的某个模块的名字： print(vgg16_true)

单独加一层，不改其他层 vgg_16_true.add_module('add_linear', nn.Linear(1000,10) vgg_16_true.classifier.add_module('add_linear', nn.Linear(1000,10)

直接修改某个模块层的某一层 print(vgg16_false) vgg_16.classifier[6] = nn.Linear(4096, 10)

https://zhuanlan.zhihu.com/p/102264694 EGGlab安装使用

https://sccn.ucsd.edu/eeglab/index.php EGGlab 官方网站，下载，tutorial，workshop。

https://blog.csdn.net/craig_cc/article/details/105530118