So, the problem is that they have per_device_batch_size = 1 by default in their implementation of Faster RCNN, and we cannot send the different value there. Quite weird, and also some other errors in the mentioned training script exist. I am currently attempting to get rid of the issues.
--- [Visit Topic](https://discuss.mxnet.apache.org/t/object-detection-finetune-f-rcnn-models/6580/4) or reply to this email to respond. You are receiving this because you enabled mailing list mode. To unsubscribe from these emails, [click here](https://discuss.mxnet.apache.org/email/unsubscribe/f99166481860f0f1e573525114efaf059fca993fc61625a5e6f0a238fe1ef9fe).
