Use custom trained ...
 
Notifications
Clear all

Question Use custom trained model with PeekingDuck

8 Posts
3 Users
0 Likes
179 Views
MarkV
Posts: 7
Member
Topic starter
(@markv)
Active Member
Joined: 2 months ago

Hello to everybody

I wanted to ask how can I use a custom trained model through PeekingDuck. I've managed to train a custom model based on the yolo v4 tiny with Roboflow and got the new weights. How should I replace the default ones of the tiny yolo v4 with my new ones?

Thanks in advance

7 Replies
Gao Hongnan
Posts: 19
Moderator
(@gao-hongnan)
Eminent Member
Joined: 4 months ago

Hi there, thanks for the question! Noted that you are using Roboflow to train the custom model.   

Let me have a short discussion with my team and get back to you! Cheers!

Reply
Gao Hongnan
Posts: 19
Moderator
(@gao-hongnan)
Eminent Member
Joined: 4 months ago

Hi, may I check if you are using RoboFlow's tutorial here?

Reply
MarkV
Posts: 7
Member
Topic starter
(@markv)
Active Member
Joined: 2 months ago

Yep this is it. I was actually trying to add a couple of classes to my yolo v4 tiny model so i've trained it with a few custom images in order to have better detection regarding these classes. But i'm not sure how to proceed with importing the trained model and use it through peeking duck instead of the default one.

Reply
Gao Hongnan
Posts: 19
Moderator
(@gao-hongnan)
Eminent Member
Joined: 4 months ago

Hi, in that case, you may have to create a custom node to inference your custom-trained model.

 

Here's a tutorial to get you started, however, if you have troubles, feel free to reach to me via messages, and we can connect further.

Reply
1 Reply
dotw
 dotw
AISG Staff
(@dotw)
Joined: 4 months ago

New Member
Posts: 1

@markv Just to explain a bit on why a custom node is required: PeekingDuck default model weights cannot be replaced, as there is a built-in check for file integrity, which will fail if the weights do not match. Hence, you need to save your custom weights in a sub-directory and write a custom node to load those weights, init the model and use that to do your custom inference. 

Reply
Page 1 / 2
Share: