train ucsd

[DengMiao55]

Hi，thanks for your work。When I train it ,I get this error。

Traceback (most recent call last): File "train.py", line 201, in main() File "train.py", line 192, in main train_ucsdped2() File "train.py", line 128, in train_ucsdped2 x_r, z, z_dist = model(x) File "/home/dl/VSST/dm/novelty-detection-master/models/base.py", line 33, in call return super(BaseModule, self).call(*args, **kwargs) File "/home/dl/anaconda3/envs/pytorch0.4/lib/python3.7/site-packages/torch/nn/modules/module.py", line 477, in call result = self.forward(*input, **kwargs) File "/home/dl/VSST/dm/novelty-detection-master/models/LSA_ucsd.py", line 189, in forward z = self.encoder(h) File "/home/dl/VSST/dm/novelty-detection-master/models/base.py", line 33, in call return super(BaseModule, self).call(*args, **kwargs) File "/home/dl/anaconda3/envs/pytorch0.4/lib/python3.7/site-packages/torch/nn/modules/module.py", line 477, in call result = self.forward(*input, **kwargs) File "/home/dl/VSST/dm/novelty-detection-master/models/LSA_ucsd.py", line 62, in forward h = self.conv(h) File "/home/dl/anaconda3/envs/pytorch0.4/lib/python3.7/site-packages/torch/nn/modules/module.py", line 477, in call result = self.forward(*input, **kwargs) File "/home/dl/anaconda3/envs/pytorch0.4/lib/python3.7/site-packages/torch/nn/modules/container.py", line 91, in forward input = module(input) File "/home/dl/VSST/dm/novelty-detection-master/models/base.py", line 33, in call return super(BaseModule, self).call(*args, **kwargs) File "/home/dl/anaconda3/envs/pytorch0.4/lib/python3.7/site-packages/torch/nn/modules/module.py", line 477, in call result = self.forward(*input, **kwargs) File "/home/dl/VSST/dm/novelty-detection-master/models/blocks_3d.py", line 133, in forward activation_fn=self._activation_fn File "/home/dl/VSST/dm/novelty-detection-master/models/blocks_3d.py", line 33, in residual_op ha = f1(ha) File "/home/dl/VSST/dm/novelty-detection-master/models/base.py", line 33, in call return super(BaseModule, self).call(*args, **kwargs) File "/home/dl/anaconda3/envs/pytorch0.4/lib/python3.7/site-packages/torch/nn/modules/module.py", line 477, in call result = self.forward(*input, **kwargs) File "/home/dl/VSST/dm/novelty-detection-master/models/layers/mconv3d.py", line 29, in forward return super(MaskedConv3d, self).forward(x) File "/home/dl/anaconda3/envs/pytorch0.4/lib/python3.7/site-packages/torch/nn/modules/conv.py", line 421, in forward self.padding, self.dilation, self.groups) RuntimeError: Expected 5-dimensional input for 5-dimensional weight [8, 1, 3, 3, 3], but got input of size [105, 690, 1, 8, 32, 32] instead

[DavideA]

Hi,

I think you are not using the correct collate function in the DataLoader. One quick way to patch this is to reshape the input tensor before feeding the network:

x_in = x_in.view(-1, *x_in.shape[2:])

Hope this helps, D

[DengMiao55]

@DavideA thank you，it‘s useful。 but I can't use the batch_size mentioned in the paper 。Do I need to use multiple GPUs？

[DavideA]

I used a single 12GB GPU. I'd say you could reduce the batch size and give it a try.

D

[DengMiao55]

@DavideA Hi，I try to recurring your result in cifar ，but I only get AUC 0.6. could you tell me more about the detail about training。 thanks

[liuem607]

Hi, I have a similar question regarding input size. If my dataloader works correctly, the input data size before patching as you mentioned is [1, 690, 1, 8, 32, 32]. If I understand correctly, this first 1 is batch_size, and 8 is 8 sequential frames, and 32*32 is resized frame shape. But what is 690?

Also, the original image size is 256384, isn't 3232 too small?

Thank you in advance!

[ColaWithIce]

@DavideA Hi，I try to recurring your result in cifar ，but I only get AUC 0.6. could you tell me more about the detail about training。 thanks

Hi, where did you get the training code?

[shen-yao]

@DavideA Hi，I try to recurring your result in cifar ，but I only get AUC 0.6. could you tell me more about the detail about training。 thanks

Hi, where did you get the training code?

Same question please

[BenedictGreen]

Thx for the great work! I got same question about the details about training. hoping for author uploading the training code.@DavideA thanks again!