Question about the training result on 4 A100 with batch size = 36

[lizhyuxi]

I trained using four A100 GUP and the total batch size is 36. After a total of 300,000 times of training, this is the result of the model:

which is quite different from the result given in your paper :

I did not change the code, what could be the cause?

[xiao-11]

How long do you train the model on 4 A100?

[swagger-coder]

@lizhyuxi I have the same issue. Do you have any idea? I find FP is very high, which means that many images with objects are incorrectly classified as having no objects.