GTA
GTA copied to clipboard
ZEKAICHEN
Learning Graph Structures with Transformer for Multivariate Time Series Anomaly Detection in IoT
Hey, Great work! I would just like to point out some potential typos I might have found while trying to run your codes, located in `gta.py` and `exp_gta_dad.py`: The naming...
Hi @ZEKAICHEN, I tried running your code on a Google Colab notebook and received the following error:  I checked what args.num_nodes contained and it printed the following:  May...
Hi,Zekai, Wondering how you evaluate the anomaly performance on datasets Wadi and Swat. Could you share this part of code? Thank you!
 为什么你的模型训练是 验证集用的是测试集里面的数据呢
感谢您提出的方法和开源代码。 但是我对比较是否公平持有怀疑，在GDN中，并没有使用[point adjustment](https://arxiv.org/pdf/1802.03903.pdf)来调整预测结果，而这一个调整策略能明显提高异常检测的准确度，所以不对GDN进行统一处理而直接使用论文中的数据来进行比较是存在不公平的，您能对此作出解释吗？
Where can I find the data to reproduce experiments?
WADI SWaT and so on. Maybe offer a small set of data for traning
Hi, when I run this code, there is a error. ``` Traceback (most recent call last): File "exp\exp_gta_dad.py", line 223, in test for i, (batch_x,batch_y,batch_x_mark,batch_y_mark,batch_label) in enumerate(test_loader): ValueError: not enough...
Hi ! Could you please specify which versions of PyTorch and PyTorch Lightning are compatible or recommended for the repository(https://github.com/chenz53/MIM-Med3D)? I have tryed various versions, but the code still reports...
