lilichu

https://github.com/roytseng-tw/Detectron.pytorch/blob/8315af319cd29b8884a7c0382c4700a96bf35bbc/lib/modeling/ResNet.py#L76 ``` def _init_modules(self): # Freeze all bn (affine) layers !!! self.apply(lambda m: freeze_params(m) if isinstance(m, mynn.AffineChannel2d) else None) ```

+1. Have anyone successfully run it?

> @alexanderpoplavsky > you can download the training set here, but it will be deleted if there're any copy rights issus > https://drive.google.com/file/d/10SGv_kbYhVLIC2hLlz2GBkHGAo0nec-3/view?usp=sharing hi, the dataset has some dirs, such...

where are the png mask files? is there a download link? thanks!

I download 'mask2014/val2014/mask_COCO_val2014_' + img_idx + 'jpg' but the code needs 'mask2014/val2014_mask_miss_' + img_idx + 'png'

感谢回复。 原作者在论文里提到用的数据量如face_cartoon和scenery_cartoon都是1w张，猜测应该是把所有的数据全放一起训练的吧，并不是分不同风格的。在评论里也没有说是分开训练的： https://github.com/SystemErrorWang/White-box-Cartoonization/issues/5#issuecomment-633866222