leapmind / mlf-sc Goto Github PK

When I run 'python main.py train cfg/sample_config.yml '
and even set batch_size==1,
the issue raised as follows:

loading images: 99%|█████████████████████████████████████████████████
loading images: 100%|█████████████████████████████████████████████████
loading images: 100%|█████████████████████████████████████████████████
█████████████| 1000/1000 [00:18<00:00, 54.16it/s]
Traceback (most recent call last):
File "main.py", line 109, in
main()
File "main.py", line 74, in main
model.train()
File "D:\research\comp_data\MLF-SC\src\models.py", line 53, in train
batch_img = p(batch_img)
File "D:\research\comp_data\MLF-SC\src\preprocessor.py", line 95, in call
x = f(x)
File "D:\anaconda\lib\site-packages\torch\nn\modules\module.py", line 541, in call
result = self.forward(*input, **kwargs)
File "D:\anaconda\lib\site-packages\torch\nn\modules\conv.py", line 345, in forward
return self.conv2d_forward(input, self.weight)
File "D:\anaconda\lib\site-packages\torch\nn\modules\conv.py", line 342, in conv2d_forward
self.padding, self.dilation, self.groups)
RuntimeError: [enforce fail at ..\c10\core\CPUAllocator.cpp:72] data. DefaultCPUAllocator: not enough memory: you tried
to allocate 16777216000 bytes. Buy new RAM!

Well, would you figure out the way to solve this issue?
Thank you in advance!

is there any reference paper？

First of all thanks for the work.

I was trying to use the test section for the prediction of anomalies from the dataset. The time taken for prediction is too long.

Is there any option for utilizing GPU with the code that can improve the execution time?

First of all, thanks for the great work. Can you tell how to get the R1 and R2 mentioned in this paper? Actually, I can just get AP and AUC in the code. There is no way to get R1 and R2. Sorry to bother you~

Hi, execuse me. I really wonder know that how you use calculate_score(neg_err, pos_err) to classify the images, especially the usage of y_score? Thank you so much.