george730 / e-resgat Goto Github PK

First, the file address identified by the code is wrong, and secondly, after the modification is completed, the corresponding file cannot be read. Please answer

您好，我将压缩包（包含上述的文件）解压在本地并不能运行，我运行fit_model文件会报错，会产生以下的错误：
Traceback (most recent call last):
File "F:\PycharmProjects\E-ResGAT-master\fit_model.py", line 180, in
fit(args)
File "F:\PycharmProjects\E-ResGAT-master\fit_model.py", line 40, in fit
edge_feat, label, adj, adj_lists, config = load_gat(path, device, binary)
File "F:\PycharmProjects\E-ResGAT-master\loader.py", line 56, in load_gat
edge_feat = np.load(path + "edge_feat_scaled.npy") # （n,f)
File "D:\Anaconda\envs\TF2.10gpu\lib\site-packages\numpy\lib\npyio.py", line 427, in load
fid = stack.enter_context(open(os_fspath(file), "rb"))
FileNotFoundError: [Errno 2] No such file or directory: 'datasets/Darknetedge_feat_scaled.npy'

然而当我将这个文件edge_feat_scaled.npy重命名为Darknetedge_feat_scaled.npy，会进一步显示报错，datasets/DarknetDarknetedge_feat_scaled.npy

我是在win系统下运行的，numpy,torch以及scikit的版本比您要求的高。
能麻烦您指导下问题所在吗？

batch response: This repository is over its data quota. Account responsible for LFS bandwidth should purchase more data packs to restore access.
Failed to fetch some objects from 'https://github.com/George730/E-ResGAT.git/info/lfs'

Dear authors,

Thank you for uploading this code. I am a beginner in anomaly detection and this has been very helpful in my research.
I have read the original article corresponding to that code. When I try to run the program, "fit_model.py", the program prompts an error, 'datasets/UNSW-NB15edge_feat_scaled.npy' could not be found, so I changed the path and the program reports the following error：

path: datasets/UNSW-NB15/
Traceback (most recent call last):
File "/usr/local/lib/python3.7/dist-packages/numpy/lib/npyio.py", line 460, in load
return pickle.load(fid, **pickle_kwargs)
_pickle.UnpicklingError: invalid load key, 'v'.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "fit_model.py", line 180, in
fit(args)
File "fit_model.py", line 40, in fit
edge_feat, label, adj, adj_lists, config = load_gat(path, device, binary)
File "/content/gdrive/MyDrive/Colab Notebooks/E-ResGAT-master/loader.py", line 58, in load_gat
edge_feat = np.load(path + "edge_feat_scaled.npy", allow_pickle=True)
File "/usr/local/lib/python3.7/dist-packages/numpy/lib/npyio.py", line 463, in load
"Failed to interpret file %s as a pickle" % repr(file))
OSError: Failed to interpret file 'datasets/UNSW-NB15/edge_feat_scaled.npy' as a pickle

And all the dataset "edge_feat_scaled.npy" is similar. So I tried to download it using git lfs, but it's still the same after downloading. How can I solve this problem?

Thanks for any help.
Regards,
LiY

Hello, I encountered the following problem when running the code. How should I solve it?

....
batch: 179 loss_train: 0.0063 acc_train: 0.9980 time: 1.9977s
batch: 180 loss_train: 0.0113 acc_train: 0.9943 time: 2.0366s
Traceback (most recent call last):
File "fit_model.py", line 180, in
fit(args)
File "fit_model.py", line 112, in fit
acc_val, loss_val, val_output = predict_(alg, model, label, loss_fn, val)
File "fit_model.py", line 142, in predict_
batch_loss = loss_fn(batch_output, label[batch_edges])
File "/home/cetc54-5/anaconda3/envs/torchIDS/lib/python3.6/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/home/cetc54-5/anaconda3/envs/torchIDS/lib/python3.6/site-packages/torch/nn/modules/loss.py", line 1152, in forward
label_smoothing=self.label_smoothing)
File "/home/cetc54-5/anaconda3/envs/torchIDS/lib/python3.6/site-packages/torch/nn/functional.py", line 2846, in cross_entropy
return torch._C._nn.cross_entropy_loss(input, target, weight, _Reduction.get_enum(reduction), ignore_index, label_smoothing)
RuntimeError: Expected floating point type for target with class probabilities, got Long

您好，我对您的研究很感兴趣，但我在复现的过程中遇到了一些问题：
fit_model.py中的命令行参数设置存在语法问题：

p.add_argument('--binary',
                   help='Perform binary or muticlass task',
                   #action='store_true',
                   type=bool,
                   default=True)

对于bool类型参数设置，无法怎么设置binary都是True，这是很基础的语法问题。
此外，我修改后模型的多分类任务上准确率能够达到98.45%，各个类别得分正常，但是绘制的散点图没有visualization.ipynb中的效果，我恳请您能提供一份完整的最终代码，我的邮箱为[email protected]，谢谢。

Can you kindly give an "in-dept" guide for running the program?

During training, the loss value is nan. Whether the processed data contains nan values.

Traceback (most recent call last):
File "D:/PyCharm/Py_project/E-ResGAT-master/fit_model.py", line 189, in
File "D:/PyCharm/Py_project/E-ResGAT-master/fit_model.py", line 114, in fit
acc_val, loss_val, val_output = predict_(alg, model, label, loss_fn, val)
File "D:/PyCharm/Py_project/E-ResGAT-master/fit_model.py", line 145, in predict_
batch_loss = loss_fn(batch_output, label[batch_edges])
File "D:\Anaconda\envs\GraphSAGE\lib\site-packages\torch\nn\modules\module.py", line 722, in _call_impl
result = self.forward(*input, **kwargs)
File "D:\Anaconda\envs\GraphSAGE\lib\site-packages\torch\nn\modules\loss.py", line 948, in forward
ignore_index=self.ignore_index, reduction=self.reduction)
File "D:\Anaconda\envs\GraphSAGE\lib\site-packages\torch\nn\functional.py", line 2422, in cross_entropy
return nll_loss(log_softmax(input, 1), target, weight, None, ignore_index, None, reduction)
File "D:\Anaconda\envs\GraphSAGE\lib\site-packages\torch\nn\functional.py", line 1591, in log_softmax
ret = input.log_softmax(dim)
IndexError: Dimension out of range (expected to be in range of [-1, 0], but got 1)

Hello, can you help me solve this problem?

Hello, I am very sorry. Since I am not very familiar with your research direction, the source file of the edge_feat_scaled.npy file of the data set cannot be downloaded. If it is convenient, I hope you can send me any edge_feat_scaled.npy file. I would like to take a look at the format of its data so that I can expand it in my direction.
this is my e-mail: [email protected]
thank you

When I was using Git to pull 'edge_feat_scaled. npy' from your repository, there was a prompt saying 'exceeded the storage limit of lfs'. Can you provide the processing code for the dataset or share this data in other ways?