Code is difficult to reproduce

[ZXW-King]

Hello, which version of tensorflow and torch are you using? I used torch1.13.0+tensorflow2.5+cuda11.2 and got an error: "FileNotFoundError: Op type not registered 'DisableCopyOnRead' in binary runn", and the model could not be loaded. After changing to tensorflow2.13.0, an error still occurred:
tensorflow.python.framework.errors_impl.UnimplementedError: Graph execution error:

Detected at node 'superpoint/pred_tower0/vgg/conv1_1/conv/Relu' defined at (most recent call last):
Node: 'superpoint/pred_tower0/vgg/conv1_1/conv/Relu'
Detected at node 'superpoint/pred_tower0/vgg/conv1_1/conv/Relu' defined at (most recent call last):
Node: 'superpoint/pred_tower0/vgg/conv1_1/conv/Relu'
2 root error(s) found.
(0) UNIMPLEMENTED: DNN library is not found.
[[{{node superpoint/pred_tower0/vgg/conv1_1/conv/Relu}}]]
[[superpoint/descriptors/_157]]
(1) UNIMPLEMENTED: DNN library is not found.
[[{{node superpoint/pred_tower0/vgg/conv1_1/conv/Relu}}]]
0 successful operations.
0 derived errors ignored.

Original stack trace for 'superpoint/pred_tower0/vgg/conv1_1/conv/Relu':
How can I solve this problem? Thank you!

[blackjacket996]

I’ve got a same problem

[cccober]

I‘ve also encountered this problem

[coffeeoreo]

I also encountered this problem

[atomishcv]

@ZXW-King I got the same problem, did you solve it