search

Junjue-Wang / LoveDA

[NeurIPS 2021] LoveDA: A Remote Sensing Land-Cover Dataset for Domain Adaptive Semantic Segmentation
344 stars 49 forks source link

Predict.py Problem #53

Open Luo-Z13 opened 1 year ago

Luo-Z13 commented 1 year ago

I download pretrained weight and use predict.py to test some images, but meet this bug, what's the problem of the fuse_layers?

File "test4/Road/LoveDA-master/Semantic_Segmentation/module/baseline/base_hrnet/_hrnet.py", line 394, in forward y = y + self.fuse_layers[i][j](x[j]) RuntimeError: The size of tensor a (500) must match the size of tensor b (504) at non-singleton dimension 3

Junjue-Wang commented 1 year ago

You maybe use your own images with incorrect input sizes.

Luo-Z13 commented 1 year ago

Thank you for your quick reply, so where should I change the input size of dataloader? Is it in ./base/loveda.py?

------------------ 原始邮件 ------------------ 发件人: "Junjue-Wang/LoveDA" @.>; 发送时间: 2022年11月15日(星期二) 下午3:39 @.>; @.**@.>; 主题: Re: [Junjue-Wang/LoveDA] Predict.py Problem (Issue #53)

You maybe use your own images with incorrect input sizes.

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

Junjue-Wang commented 1 year ago

Yes. I can try it.