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.

Sign up for GitHub

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

Jump to bottom

about mean_value and scale #64

Open

JPBoost opened this issue Jan 28, 2019 · 1 comment

JPBoost commented Jan 28, 2019

Hi Robert, thanks for your paper and codes, great work!

I noticed that the in train_voc.py and train_coco.py, there are data transform_param:
scale：0.017
mean_value：103.94，116.78，123.68

and in MobileNet_SSD, the transform_param seems to be:
scale: 0.007843
mean_value: 127.5, 127.5, 127.5

both transform_params are ok for train and test as I have trained MobileNet_SSD and Pelee for my own data (to detect head and have only 1 class)

So I have two questions here:

where these transform_params come from?
how did they impact mAP and FPS?

looking forward for your reply, thank you!

zhwis commented Sep 18, 2019

@vincent8877, How to use your own data to train the accuracy of the model in pelee？I also have only one class

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment