How do I train a detector using orginal batchsize…… about borderdet

[FateScript]

When I use the --4 gpus, the training setting is automatically changed, i.e., iterations are changed from 90000 to 180000, as same as the smaller lr and batchsize.

If I have enough GPU memory, I can simply use the orginal training setting with larger batchsize like 16.

Default config is used for 8-GPU, if only 4 GPUs is detected while training, the config will change automatically. You don't need to change the config.py manually, If 8 GPUs is detected, the config will not change.

[Xudangliatiger]

Thanks. Actually I want to use the 8 GPUs' config for 4 GPU, since that I had to design some experiments about normalization layers.