Closed Naykira closed 2 years ago
When I used hg_refined_200_tiny occurred same error.
but hg_refined_200 works well.
Would it be possible to please share the changes, if any, you made to the code to get the hg_refined_200 working? Because I am facing the above issue with both tiny and normal hg implementation!
I just changed it to config.cfg and predict.load_model(load = 'hg_refined_200')
I can get it working now but it's leading to wrong detection of body keypoints. Maybe it's because the original images are resized to accommodate the network. But thanks a lot! Do you any idea why the batchnorm error comes up and how it can probably be solved?
Hello, thanks a lot for the tensorflow implementation of hourglass architecture. I have been trying to test the models and I am running into the below error during inference;
Tensor name "BatchNorm/beta/RMSProp" not found in checkpoint files /mnt//hourglass/hourglass/RegHG/hg_refined_200.index
Has anyone else encountered this issue? I would be grateful for any suggestions regarding how this could be solved. Thanks!