[prompt_segmenting.py] RuntimeError: indices should be either on cpu or on the same device as the indexed tensor (cpu)

[cs-mshah]

I am facing I selected 3 points and set them to input_point = np.array(selected_points)

I'm getting the SAM masks:

I tried adding .cuda() (to mask_pooling_prototype.unsqueeze(0)) and removing .cuda() (in cluster_centers[cluster_mask, :].cuda()) but that didn't help.

[Jumpat]

Hi, you may need to set the cluster_mask to cuda.

[CrescentRosexx]

Hi, you may need to set the cluster_mask to cuda.

我也遇到了一样的问题，有时候可以正常运行，但是修改input point或者换场景都有可能出现这个报错

[cs-mshah]

Yes, I encountered it while modifying the input point as well.

[KevinCDJ]

当我修改input_point之后遇到了同样的问题，但是我找不到问题出在哪里

[KevinCDJ]

我认为我解决了这个问题，当我尝试输出mask_pooling_prototype、cluster_centers、cluster_mask这三个张量运行的设备时，发现cluster_centers运行在cpu上（未修改过代码），然后我在79行报错（我自己是75行）后加入.cuda也不能解决，后发现在计算cluster_centers时，下面加一行 cluster_centers = cluster_centers.cuda()，即可解决