在遇到池化层的时候报错

[ZZHHogan]

cnn.conv3
conv: batch_norm_blob2
conv6 was added to layers
144541320:conv_blob6 was added to blobs
cnn.relu3
relu4 was added to layers
144541720:relu_blob4 was added to blobs
cnn.convproject3
conv: relu_blob4
conv7 was added to layers
144541560:conv_blob7 was added to blobs
cnn.pooling2
max_pool2 was added to layers
144541480:max_pool_blob2 was added to blobs
Traceback (most recent call last):
File "E:/pro/PytorchToCaffe-master/example/testConvert.py", line 229, in
pytorch_to_caffe.trans_net(net,input,name)
File "E:\pro\PytorchToCaffe-master\pytorch_to_caffe.py", line 789, in trans_net
out = net.forward(input_var)
File "E:/pro/PytorchToCaffe-master/example/testConvert.py", line 116, in forward
conv = self.cnn(input)
File "E:\anaconda\lib\site-packages\torch\nn\modules\module.py", line 493, in call
result = self.forward(*input, **kwargs)
File "E:\anaconda\lib\site-packages\torch\nn\modules\container.py", line 92, in forward
input = module(input)
File "E:\anaconda\lib\site-packages\torch\nn\modules\module.py", line 493, in call
result = self.forward(*input, **kwargs)
File "E:\anaconda\lib\site-packages\torch\nn\modules\pooling.py", line 146, in forward
self.return_indices)
File "E:\pro\PytorchToCaffe-master\pytorch_to_caffe.py", line 693, in call
out=self.obj(self.raw,*args,**kwargs)
File "E:\pro\PytorchToCaffe-master\pytorch_to_caffe.py", line 186, in _max_pool2d
_pool('max',raw,input, x, kernel_size, stride, padding,ceil_mode)
File "E:\pro\PytorchToCaffe-master\pytorch_to_caffe.py", line 170, in _pool
pad=padding, type=type.upper() , ceil_mode = ceil_mode)
File "E:\pro\PytorchToCaffe-master\Caffe\layer_param.py", line 117, in pool_param
pool_param.stride=pair_process(stride)
File "E:\pro\PytorchToCaffe-master\Caffe\layer_param.py", line 10, in pair_process
raise ValueError("number in item {} must be the same".format(item))
ValueError: number in item (2, 1) must be the same

[yanty123]

请问兄弟拟解决这个问题了吗，好像是这个avg要求输入图像的长宽一致

[yanzhou-li]

cnn.conv3
conv: batch_norm_blob2
conv6 was added to layers
144541320:conv_blob6 was added to blobs
cnn.relu3
relu4 was added to layers
144541720:relu_blob4 was added to blobs
cnn.convproject3
conv: relu_blob4
conv7 was added to layers
144541560:conv_blob7 was added to blobs
cnn.pooling2
max_pool2 was added to layers
144541480:max_pool_blob2 was added to blobs
Traceback (most recent call last):
File "E:/pro/PytorchToCaffe-master/example/testConvert.py", line 229, in
pytorch_to_caffe.trans_net(net,input,name)
File "E:\pro\PytorchToCaffe-master\pytorch_to_caffe.py", line 789, in trans_net
out = net.forward(input_var)
File "E:/pro/PytorchToCaffe-master/example/testConvert.py", line 116, in forward
conv = self.cnn(input)
File "E:\anaconda\lib\site-packages\torch\nn\modules\module.py", line 493, in call
result = self.forward(*input, **kwargs)
File "E:\anaconda\lib\site-packages\torch\nn\modules\container.py", line 92, in forward
input = module(input)
File "E:\anaconda\lib\site-packages\torch\nn\modules\module.py", line 493, in call
result = self.forward(*input, **kwargs)
File "E:\anaconda\lib\site-packages\torch\nn\modules\pooling.py", line 146, in forward
self.return_indices)
File "E:\pro\PytorchToCaffe-master\pytorch_to_caffe.py", line 693, in call
out=self.obj(self.raw,*args,**kwargs)
File "E:\pro\PytorchToCaffe-master\pytorch_to_caffe.py", line 186, in _max_pool2d
_pool('max',raw,input, x, kernel_size, stride, padding,ceil_mode)
File "E:\pro\PytorchToCaffe-master\pytorch_to_caffe.py", line 170, in _pool
pad=padding, type=type.upper() , ceil_mode = ceil_mode)
File "E:\pro\PytorchToCaffe-master\Caffe\layer_param.py", line 117, in pool_param
pool_param.stride=pair_process(stride)
File "E:\pro\PytorchToCaffe-master\Caffe\layer_param.py", line 10, in pair_process
raise ValueError("number in item {} must be the same".format(item))
ValueError: number in item (2, 1) must be the same

@hhhhhhogan @yanty123 ,Hi ,I also have met this problem.Have you solved the problem?Can you give me some advise. thank you