add ghostbottleneck module into yolov5s.yaml

[polarbearwy]

❔Question

when i add ghostbottleneck module into yolov5s stucture, issue shows.

Additional context

RuntimeError: invalid argument 0: Sizes of tensors must match except in dimension 1. Got 34 and 62 in dimension 2 at /pytorch/aten/src/TH/genric/THTensor.cpp:612

[github-actions]

Hello @polarbearwy, thank you for your interest in our work! Please visit our Custom Training Tutorial to get started, and see our Jupyter Notebook , Docker Image, and Google Cloud Quickstart Guide for example environments.

If this is a bug report, please provide screenshots and minimum viable code to reproduce your issue, otherwise we can not help you.

If this is a custom model or data training question, please note that Ultralytics does not provide free personal support. As a leader in vision ML and AI, we do offer professional consulting, from simple expert advice up to delivery of fully customized, end-to-end production solutions for our clients, such as:

• Cloud-based AI systems operating on hundreds of HD video streams in realtime.
• Edge AI integrated into custom iOS and Android apps for realtime 30 FPS video inference.
• Custom data training, hyperparameter evolution, and model exportation to any destination.

For more information please visit https://www.ultralytics.com.

[glenn-jocher]

@polarbearwy you need to add custom modules you want to use to yolo.py to let it know to expect them. You can do that here:

yolov5/models/yolo.py

Lines 175 to 198 in 1e94bcf

	if m in [nn.Conv2d, Conv, Bottleneck, SPP, DWConv, MixConv2d, Focus, CrossConv, BottleneckCSP, C3]:
	c1, c2 = ch[f], args[0]
	# Normal
	# if i > 0 and args[0] != no: # channel expansion factor
	# ex = 1.75 # exponential (default 2.0)
	# e = math.log(c2 / ch[1]) / math.log(2)
	# c2 = int(ch[1] * ex ** e)
	# if m != Focus:
	c2 = make_divisible(c2 * gw, 8) if c2 != no else c2
	# Experimental
	# if i > 0 and args[0] != no: # channel expansion factor
	# ex = 1 + gw # exponential (default 2.0)
	# ch1 = 32 # ch[1]
	# e = math.log(c2 / ch1) / math.log(2) # level 1-n
	# c2 = int(ch1 * ex ** e)
	# if m != Focus:
	# c2 = make_divisible(c2, 8) if c2 != no else c2
	args = [c1, c2, *args[1:]]
	if m in [BottleneckCSP, C3]:
	args.insert(2, n)
	n = 1

[polarbearwy]

I have already add custom modules to yolo.py, but there are still problems

[mary-0830]

I have same question.

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[orangezishuai]

I have already add custom modules to yolo.py, but there are still problems

hello，are you slove this issue？thanks！

[glenn-jocher]

@polarbearwy @python-faker there was a recent PR #1176 for a bug fix in GhostConv() modules. Please git pull and try again, and if the error repeats please submit a full Bug Report with code to reproduce. Thank you.

[orangezishuai]

@polarbearwy @python-faker there was a recent PR #1176 for a bug fix in GhostConv() modules. Please git pull and try again, and if the error repeats please submit a full Bug Report with code to reproduce. Thank you.

thank you very much，I solved the problem through your method.
The reason for the problem is the lack of input parameters. My comparison is as follows

before
class GhostConv(nn.Module): # Ghost Convolution https://github.com/huawei-noah/ghostnet def __init__(self, c1, c2, k=1, s=1, g=1, act=True): # ch_in, ch_out, kernel, stride, groups super(GhostConv, self).__init__() c_ = c2 // 2 # hidden channels # self.cv1 = Conv(c1, c_, k, s, g, act) # self.cv2 = Conv(c_, c_, 5, 1, c_, act)
after
class GhostConv(nn.Module): # Ghost Convolution https://github.com/huawei-noah/ghostnet def __init__(self, c1, c2, k=1, s=1, g=1, act=True): # ch_in, ch_out, kernel, stride, groups super(GhostConv, self).__init__() c_ = c2 // 2 # hidden channels self.cv1 = Conv(c1, c_, k, s, None, g, act) self.cv2 = Conv(c_, c_, 5, 1, None, c_, act)

[glenn-jocher]

Great!

[lucasjinreal]

@python-faker how's the ghostconv performance after you changed?

[Ankur-singh]

@glenn-jocher & @python-faker what did you use for weights? I tried using yolov5s.pt but most of the weights are simply ignored. In the output it showed Transferred 146/553 items from yolov5s.pt
It would be great if someone can share the pretrained weights.

[glenn-jocher]

@Ankur-singh all of our COCO models are trained from scratch with --weights ''