Issue with training CIFAR 3pxnet model about 3pxnet HOT 4 CLOSED

Hi, could you tell me which dataset are you training on? And did you make any changes to any source code? Thanks

from 3pxnet.

I used the command:

make CIFAR =1 train

no changes have been made to the source code

from 3pxnet.

Hi ando,
Sorry for the late reply. It should be fixed now. The problem is that previously the Makefile chose a wrong dataset.

from 3pxnet.

Error during: make CIFAR =1 train

Terminal produces error on windows then continues with permuting.

Epoch: [24] Loss 0.2136 (0.2697) Prec@1 94.886 (92.071) Prec@5 100.000 (99.262)
Permuting BinarizeConv2d(3, 128, kernel_size=(3, 3), stride=(1, 1), bias=False)
Traceback (most recent call last):
File "C:\Users\andre\source\repos\3pxnet\3pxnet-training\utils_own.py", line 242, in permute_all_weights_once
mod.permute_list[0] = one_time_permute(weight_flat, mod.thres, pack=pack, weight_grad_gpu = grad_flat)
File "C:\Users\andre\AppData\Local\Programs\Python\Python39\lib\site-packages\torch\nn\modules\module.py", line 1130, in getattr
raise AttributeError("'{}' object has no attribute '{}'".format(
AttributeError: 'BinarizeConv2d' object has no attribute 'thres'

from 3pxnet.