Cross-Scale-Non-Local-Attention
Cross-Scale-Non-Local-Attention copied to clipboard
PyTorch code for our paper "Image Super-Resolution with Cross-Scale Non-Local Attention and Exhaustive Self-Exemplars Mining" (CVPR2020).
I found that it is hard to use those code based on EDSR.Because I don`t know how to set the directory and other operation. For example,I put my dataset in...
> 多谢~按照你给的参数已经解决了 can you help me?  _Originally posted by @Ralapfranco in https://github.com/SHI-Labs/Cross-Scale-Non-Local-Attention/issues/1#issuecomment-1168128390_
您好,您在训练模型的时候需要多少内存才能够保证训练能够正常完成呢?现在我遇到的问题是:训练能够正常启动,但是随着训练的进行,内存占用越来越大,然后会导致内存溢出。请问这是因为内存不够还是其他原因?(我只是用了您的模型,数据处理等都是用的自己的)。
我现在在8卡的GPU上,加了apex进行混合精度训练。同样的参数,scale=2时可以达到比您高的PSNR,但是scale=4却只能跑到您的结果的99.2%。我看源码中对scale=4进行了特殊处理,有如下疑问: `if scale == 4: min_size = 3600` 这里的min_size指的是一张卡的memory吗?3600指的是什么指标?
Hi, I train your model on DIV2K dataset and test Manga109 dataset. In your paper, you got at least 39dB result on Manga109 dataset. However my model obtained 27-28dB on...
Hi! i dont konw how to get the model.latest when train the model.IS it average of all models of training?
Set5数据测试
我在测试Set5数据时,出来的结果和你们的不一样,我这边是划分为scale*scale块状且分辨率很低的图。
Where do i have to put the models and the dataset folders?
On trying to train the model for X3 scale, I am getting a runtime error: Args: --chop --batch_size 4 --model CSNLN --patch_size 96 --save CSNLN_x2 --n_feats 128 --depth 12 --data_train...
nii格式或者img格式加载会直接报错