Output nodes before 5D Reshape

[PiyalGeorge]

Can you specify the output nodes before 5D Reshape?
I'm using a 320X320 yolov5s model from 3.1 release.

Or

In other way:
I converted the yolov5s model to onnx model following this .
Then i simplified the onnx model using this . I tried this in yolov5s - release number :3.1

I wish to know "the output nodes before 5D Reshape". How can i get these output nodes? Kindly help me to know these output nodes.

[github-actions]

👋 Hello @PiyalGeorge, thank you for your interest in 🚀 YOLOv5! Please visit our ⭐️ Tutorials to get started, where you can find quickstart guides for simple tasks like Custom Data Training all the way to advanced concepts like Hyperparameter Evolution.

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 training ❓ Question, please provide as much information as possible, including dataset images, training logs, screenshots, and a public link to online W&B logging if available.

For business inquiries or professional support requests please visit https://www.ultralytics.com or email Glenn Jocher at glenn.jocher@ultralytics.com.

Requirements

Python 3.8 or later with all requirements.txt dependencies installed, including torch>=1.7. To install run:

$ pip install -r requirements.txt

Environments

YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):

• Google Colab and Kaggle notebooks with free GPU:
• Google Cloud Deep Learning VM. See GCP Quickstart Guide
• Amazon Deep Learning AMI. See AWS Quickstart Guide
• Docker Image. See Docker Quickstart Guide

Status

If this badge is green, all YOLOv5 GitHub Actions Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training (train.py), testing (test.py), inference (detect.py) and export (export.py) on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

[glenn-jocher]

@PiyalGeorge I'm not sure exactly what you mean by the output nodes before 5D reshape, though the 5D reshape is in the Detect layer, so what you are looking for is probably there.

yolov5/models/yolo.py

Lines 24 to 58 in 5f7d39f

	class Detect(nn.Module):
	stride = None # strides computed during build
	export = False # onnx export
	def __init__(self, nc=80, anchors=(), ch=()): # detection layer
	super(Detect, self).__init__()
	self.nc = nc # number of classes
	self.no = nc + 5 # number of outputs per anchor
	self.nl = len(anchors) # number of detection layers
	self.na = len(anchors[0]) // 2 # number of anchors
	self.grid = [torch.zeros(1)] * self.nl # init grid
	a = torch.tensor(anchors).float().view(self.nl, -1, 2)
	self.register_buffer('anchors', a) # shape(nl,na,2)
	self.register_buffer('anchor_grid', a.clone().view(self.nl, 1, -1, 1, 1, 2)) # shape(nl,1,na,1,1,2)
	self.m = nn.ModuleList(nn.Conv2d(x, self.no * self.na, 1) for x in ch) # output conv
	def forward(self, x):
	# x = x.copy() # for profiling
	z = [] # inference output
	self.training |= self.export
	for i in range(self.nl):
	x[i] = self.m[i](x[i]) # conv
	bs, _, ny, nx = x[i].shape # x(bs,255,20,20) to x(bs,3,20,20,85)
	x[i] = x[i].view(bs, self.na, self.no, ny, nx).permute(0, 1, 3, 4, 2).contiguous()
	if not self.training: # inference
	if self.grid[i].shape[2:4] != x[i].shape[2:4]:
	self.grid[i] = self._make_grid(nx, ny).to(x[i].device)
	y = x[i].sigmoid()
	y[..., 0:2] = (y[..., 0:2] * 2. - 0.5 + self.grid[i]) * self.stride[i] # xy
	y[..., 2:4] = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i] # wh
	z.append(y.view(bs, -1, self.no))
	return x if self.training else (torch.cat(z, 1), x)

On a side note, onnx-simplifier is now integrated with YOLOv5 export via PR #2815, you can access it like this in the latest code:

python export.py --simplify

[PiyalGeorge]

@glenn-jocher , Thanks for your reply.
By "the output nodes before 5D Reshape" - I'm expecting the three output nodes like 742, 762, 782.
These node number(eg:742, 762, etc) seems different in each release of yolov5s.
What i want is, for the release number - 3.1 , yolov5s, 320X320 (or 416X416). Hope you got what i had in mind.

[glenn-jocher]

@PiyalGeorge these operations are in the Detect() layer before merging into a single output:

yolov5/models/yolo.py

Lines 24 to 58 in 5f7d39f

	class Detect(nn.Module):
	stride = None # strides computed during build
	export = False # onnx export
	def __init__(self, nc=80, anchors=(), ch=()): # detection layer
	super(Detect, self).__init__()
	self.nc = nc # number of classes
	self.no = nc + 5 # number of outputs per anchor
	self.nl = len(anchors) # number of detection layers
	self.na = len(anchors[0]) // 2 # number of anchors
	self.grid = [torch.zeros(1)] * self.nl # init grid
	a = torch.tensor(anchors).float().view(self.nl, -1, 2)
	self.register_buffer('anchors', a) # shape(nl,na,2)
	self.register_buffer('anchor_grid', a.clone().view(self.nl, 1, -1, 1, 1, 2)) # shape(nl,1,na,1,1,2)
	self.m = nn.ModuleList(nn.Conv2d(x, self.no * self.na, 1) for x in ch) # output conv
	def forward(self, x):
	# x = x.copy() # for profiling
	z = [] # inference output
	self.training |= self.export
	for i in range(self.nl):
	x[i] = self.m[i](x[i]) # conv
	bs, _, ny, nx = x[i].shape # x(bs,255,20,20) to x(bs,3,20,20,85)
	x[i] = x[i].view(bs, self.na, self.no, ny, nx).permute(0, 1, 3, 4, 2).contiguous()
	if not self.training: # inference
	if self.grid[i].shape[2:4] != x[i].shape[2:4]:
	self.grid[i] = self._make_grid(nx, ny).to(x[i].device)
	y = x[i].sigmoid()
	y[..., 0:2] = (y[..., 0:2] * 2. - 0.5 + self.grid[i]) * self.stride[i] # xy
	y[..., 2:4] = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i] # wh
	z.append(y.view(bs, -1, self.no))
	return x if self.training else (torch.cat(z, 1), x)

[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.

[PiyalGeorge]

thanks @glenn-jocher ,

The reason i asked the question is because of this - https://developer.qualcomm.com/docs/snpe/revision_history.html
As per the qualcomm, they have added support for yolov5. So i was trying to figure it out. But still no luck.

[glenn-jocher]

@PiyalGeorge that's awesome that Qualcomm has added support for YOLOv5 models!! I didn't know about that. I wonder if they have more documentation?

[PiyalGeorge]

My ideas are from these links:-

1. https://developer.qualcomm.com/comment/reply/68483#comment-form
2. https://docs.ultralytics.com/yolov5/tutorials/model_export
3. https://developer.qualcomm.com/docs/snpe/model_conv_tensorflow.html
4. Pytorch->ONNX->TF-> SNPE Error for batchnorm variance onnx/onnx-tensorflow#749

In the qualcomm's SNPE, the final model format we want is a .dlc file. Since there is no direct method to convert .pt file to .dlc, we need to convert the .pt file to onnx then to .dlc (that's the whole idea i get from first link).
So first need to convert a .pt file to onnx, using 2nd link OR onnx-simplifier mentioned in 1st link.
Once we get onnx we need the out_node to convert it to .dlc like mentioned in link 3, 4. Used netron viewer to view the onnx model and there are three output nodes.

[glenn-jocher]

@PiyalGeorge you can skip the Detect() layer concatenation and grid op with the --train mode flag:

python models/export.py --train

Make sure you use the latest master, we recently pushed a fix for this.

[github-actions]

👋 Hello, this issue has been automatically marked as stale because it has not had recent activity. Please note it will be closed if no further activity occurs.

Access additional YOLOv5 🚀 resources:

• Wiki – https://github.com/ultralytics/yolov5/wiki
• Tutorials – https://docs.ultralytics.com/yolov5
• Docs – https://docs.ultralytics.com

Access additional Ultralytics ⚡ resources:

• Ultralytics HUB – https://ultralytics.com
• Vision API – https://ultralytics.com/yolov5
• About Us – https://ultralytics.com/about
• Join Our Team – https://ultralytics.com/work
• Contact Us – https://ultralytics.com/contact

Feel free to inform us of any other issues you discover or feature requests that come to mind in the future. Pull Requests (PRs) are also always welcomed!

Thank you for your contributions to YOLOv5 🚀 and Vision AI ⭐!

[evdoks]

@PiyalGeorge were you able to export an already trained YOLOv5 model to DLC?
My understanding is that export with --train flag results in ONNX representation without the detection layer and thus cannot be used for detection. What would be the purpose of such an export or am I missing something?

[PiyalGeorge]

Sorry to disappoint, i couldnt figure it out. Its possible, but i was busy.