Hanqiu Jiang

Further debugging reveals that it works if I set `exclude_managed_identity_credential` to `True` in `DefaultAzureCredential` in ``dvc/lib/python3.8/site-packages/dvc_objects/fs/implementations/azure.py: ``` login_info["credential"] = DefaultAzureCredential( exclude_interactive_browser_credential=False, exclude_environment_credential=config.get( "exclude_environment_credential", False ), exclude_visual_studio_code_credential=config.get( "exclude_visual_studio_code_credential", False ), exclude_shared_token_cache_credential=config.get(...

Maybe you could try this scipy==0.19.1 version? Unfortunately, I currently can't test without launching a AWS instance.

Yes, you are correct.

Hi, thank you! The `num_classes` parameter refers to the original ENet and has to be fixed at 12, so it can be loaded from the checkpoint file. This layer is...

Hi. The loading worked for other people, so you might want to check your tensorflow version. It should be around 1.2 -1.3, perhaps you have a newer version which is...

I was also surprised that mean shift is quite slow. What you can do is, increase the number of threads in the mean shift and reduce the tolerance to stop...