-
Notifications
You must be signed in to change notification settings - Fork 315
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
PaddleViT用于语义分割数据集加载问题 #169
Labels
Comments
Hello, 感谢使用PaddleViT,您的问题我们已经收到。 @wutianyiRosun 能否尽快看一下并处理这个问题?Thanks! |
@xiaoguoguo2018 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
您好!我在使用PaddleViT语义分割用于自己的数据集的时候遇到了问题,希望您可以抽空帮我解决一下
1.使用自己的数据集用于语义分割,或者Cityscapes,ADE20K等数据集,lable标签载入应该是彩色的图片,还是已经将颜色(如0,255,255)转换为对应类别(如类别1,)后的黑色图片呢?目前我没有在代码中找到label标签颜色和类别转换的代码。
2.为什么使用数据增强策略,在模型是val的时候,图片进行了数据增强,而label没有对应的也进行相应的变换。而是直接open图片?如在semantic_segementation/src/datasets/ade.py和vaihingen.py
if self.mode == 'val':
img, _ = self.transforms(img=image_path) # --> img进行相应的缩放,裁剪等数据增强策略
label = np.asarray(Image.open(label_path)) #--> label为什么不做对应的数据增强,如缩放,裁剪
label = label - 1
label = label[np.newaxis, :, :]
return img, label
期待开发者团队您的回复! @xperzy
The text was updated successfully, but these errors were encountered: