Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

How to solve this problem #334

Open
huanai opened this issue Nov 25, 2020 · 4 comments
Open

How to solve this problem #334

huanai opened this issue Nov 25, 2020 · 4 comments

Comments

@huanai
Copy link

huanai commented Nov 25, 2020

IMG_3718(20201125-201631)

@huanai huanai closed this as completed Nov 25, 2020
@huanai huanai reopened this Nov 25, 2020
@a28293971
Copy link

check your anchors size and class num if you tain your custom dataset.
please note that, there are many const value in the class Yolov4 and function train, you also need to change them to the correct value which fit for your dataset.
e.g. anchors, image_size or n_classes...

@Mulbetty
Copy link

Mulbetty commented Dec 5, 2020

Hi, how do you solve the problem? Thank you

@a28293971
Copy link

Hi, how do you solve the problem? Thank you

As I said above, you need to understand the theory of Yolov4 first, and then go to the source code to modify the parts that need to be fitted for your dataset size.

@ErlingLie
Copy link

You should try this solution #138 .

You need to change the classes values in the cfg so that it matches your number of classes, and also the number of filters in the previous layer such that it equals 3*(5 + num classes) instead of 255.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants