Skip to content
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

error in ms_deformable_im2col_cuda: no kernel image is available for execution on the device #5

Open
huiyang0613 opened this issue Nov 3, 2023 · 6 comments

Comments

@huiyang0613
Copy link

你好,我按照github上您发布的步骤进行测试,输入图片是椅子,然后实例提示,和关键点提示都是参照提前预定好的,但是测试时"pred_logits"这个变量中出现了大量的inf值导致无法输出测试结果。同时在我测试的时候还伴随着error in ms_deformable_im2col_cuda: no kernel image is available for execution on the device这行警告,我想知道这二者之间是否有联系以及出现上述问题该如何解决呢?
如果能得到您的回复真是太感谢了~

@huiyang0613
Copy link
Author

按照您之前提供的parser.add_argument("--box_threshold", type=float, default=0.1, help="box threshold") 边界框阈值,我得不到了错误警告,因为阈值过大将所有可能的点都过滤完了,然后我将"--box_threshold参数修改为--box_threshold=0.01得到了以下输出
image
很明显,这个结果得到了太多多余的边界框和关键点,请问--box_threshold这个参数该怎么设置呢

@huiyang0613
Copy link
Author

image
我按照您提供的关键点先验自己设定了一个易拉罐的类别,但是如图它表现的并不好,请问unipose是否适用于 易拉罐,笔记本电脑,碗,相机这种生活物品呢

@yangjie-cv
Copy link
Collaborator

你好,我按照github上您发布的步骤进行测试,输入图片是椅子,然后实例提示,和关键点提示都是参照提前预定好的,但是测试时"pred_logits"这个变量中出现了大量的inf值导致无法输出测试结果。同时在我测试的时候还伴随着error in ms_deformable_im2col_cuda: no kernel image is available for execution on the device这行警告,我想知道这二者之间是否有联系以及出现上述问题该如何解决呢? 如果能得到您的回复真是太感谢了~

error in ms_deformable_im2col_cuda: no kernel image is available for execution on the device, indicating that the network is not running deformable attention. You need to follow the installation guide to configure the CUDA operator.

@yangjie-cv
Copy link
Collaborator

按照您之前提供的parser.add_argument("--box_threshold", type=float, default=0.1, help="box threshold") 边界框阈值,我得不到了错误警告,因为阈值过大将所有可能的点都过滤完了,然后我将"--box_threshold参数修改为--box_threshold=0.01得到了以下输出 image 很明显,这个结果得到了太多多余的边界框和关键点,请问--box_threshold这个参数该怎么设置呢

You should try different thresholds separately to get satisfactory results. In order to remove multiple outputs of the same object, you can set it through the iou threshold.

@yangjie-cv
Copy link
Collaborator

image 我按照您提供的关键点先验自己设定了一个易拉罐的类别,但是如图它表现的并不好,请问unipose是否适用于 易拉罐,笔记本电脑,碗,相机这种生活物品呢

These objects belong to the unseen category (including objects and keypoints). The model does not know what keypoints you want to detect. You need to define the keypoints to be detected through the visual prompt.

@huiyang0613
Copy link
Author

image 我按照您提供的关键点先验自己设定了一个易拉罐的类别,但是如图它表现的并不好,请问unipose是否适用于 易拉罐,笔记本电脑,碗,相机这种生活物品呢

These objects belong to the unseen category (including objects and keypoints). The model does not know what keypoints you want to detect. You need to define the keypoints to be detected through the visual prompt.

请问大概什么时候会发布视觉提示的预训练模型呢非常期待

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants