search

ainrichman / Peppa-Facial-Landmark-PyTorch

Facial Landmark Detection based on PyTorch
Apache License 2.0
211 stars 56 forks source link

speed up the training #16

Open vgg4resnet opened 3 years ago

vgg4resnet commented 3 years ago

hi,I found it a long time to train one epoch, is there any way to speed up，I wonder which part of the augment code take so much time,？Have you got a QQ ?or do we have the QQ group?

ainrichman commented 3 years ago

Hello, the runtime transformation of the images cost most of the training time. This is to be ameliorated. QQ 183526402。