When one block has no channels to preserve?

[zssjh]

hello，after I run the prune_model_imagenet.py，when i == 2 in following code, the pruned_cfg[i]=0，it means that the block has no channels to keep.

 for i in range(len(pruned_cfg)):
        pruned_cfg[i] = masks[counter].sum().long().item() 
        counter += 1
    model = models.__dict__[args.arch](args.num_classes, args.expanded_inchannel, args.multiplier, pruned_cfg)

Then, the error occurs:

Traceback (most recent call last):
File "script/prune_model_imagenet.py", line 82, in
model = models.dict[args.arch](args.num_classes, args.expanded_inchannel, args.multiplier, pruned_cfg)
File "/pruning-from-scratch/models/mobilenet.py", line 227, in mobilenet_v2
return MobileNetV2(num_class, in_channel, multiplier, cfg)
File "/pruning-from-scratch/models/mobilenet.py", line 189, in init
self.features.append(InvertedResidual(inp, oup, hid, stride))
File "/pruning-from-scratch/models/mobilenet.py", line 159, in init
self.conv = InvertedBlock(inp, oup, hid, stride)
File "/pruning-from-scratch/models/mobilenet.py", line 117, in init
self.conv1 = nn.Conv2d(inp, hid, 1, 1, 0, bias=False)
File "python3.7/site-packages/torch/nn/modules/conv.py", line 338, in init
False, pair(0), groups, bias, padding_mode)
File "python3.7/site-packages/torch/nn/modules/conv.py", line 53, in init
self.reset_parameters()
File "python3.7/site-packages/torch/nn/modules/conv.py", line 56, in reset_parameters
init.kaiming_uniform(self.weight, a=math.sqrt(5))
File "python3.7/site-packages/torch/nn/init.py", line 322, in kaiming_uniform_
fan = _calculate_correct_fan(tensor, mode)
File "python3.7/site-packages/torch/nn/init.py", line 291, in _calculate_correct_fan
fan_in, fan_out = _calculate_fan_in_and_fan_out(tensor)
File "python3.7/site-packages/torch/nn/init.py", line 223, in _calculate_fan_in_and_fan_out
receptive_field_size = tensor[0][0].numel()
IndexError: index 0 is out of bounds for dimension 0 with size 0

How do I solve the problem? Thanks so much~