Pytorch_Retinaface
Pytorch_Retinaface copied to clipboard
R50 train parameter
Is the R50 result displayed on the page consistent with the training parameters of the mxnet version? I see that the training scale is 840 * 840 in R50‘s config.
Same question. And the initial learning rate in the paper is 1e-2, while 1e-3 in this repo here. Have you checked if we can enlarge the lr while reducing the training epochs?
@dianxin556 have you ever tried one 640x640 trained model? My device is not allowed to train 24 images for 840x840 :(
Hi, I've trained one 640x640 baseline, the result is ==================== Results ==================== Easy Val AP: 0.9270121584603086 Medium Val AP: 0.9234432537094537 Hard Val AP: 0.884572269484029
can you share your training config parameters?