Unable to get a Vulkan tensor using to('vulkan')

[skyline75489]

Am I doing something wrong, or this is expected?

>>> torch.is_vulkan_available()
True
>>> a = torch.rand(1,3)
>>> a.to('vulkan')
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "C:\Users\skyline\.conda\envs\pytorch-build-py37\lib\site-packages\torch\_tensor.py", line 203, in __repr__
    return torch._tensor_str._str(self)
  File "C:\Users\skyline\.conda\envs\pytorch-build-py37\lib\site-packages\torch\_tensor_str.py", line 406, in _str
    return _str_intern(self)
  File "C:\Users\skyline\.conda\envs\pytorch-build-py37\lib\site-packages\torch\_tensor_str.py", line 381, in _str_intern
    tensor_str = _tensor_str(self, indent)
  File "C:\Users\skyline\.conda\envs\pytorch-build-py37\lib\site-packages\torch\_tensor_str.py", line 242, in _tensor_str
    formatter = _Formatter(get_summarized_data(self) if summarize else self)
  File "C:\Users\skyline\.conda\envs\pytorch-build-py37\lib\site-packages\torch\_tensor_str.py", line 90, in __init__
    nonzero_finite_vals = torch.masked_select(tensor_view, torch.isfinite(tensor_view) & tensor_view.ne(0))
NotImplementedError: Could not run 'aten::abs.out' with arguments from the 'Vulkan' backend. This could be because the operator doesn't exist for this backend, or was omitted during the selective/custom build process (if using custom build). If you are a Facebook employee using PyTorch on mobile, please visit https://fburl.com/ptmfixes for possible resolutions. 'aten::abs.out' is only available for these backends: [CPU, BackendSelect, Named, InplaceOrView, AutogradOther, AutogradCPU, AutogradCUDA, AutogradXLA, UNKNOWN_TENSOR_TYPE_ID, AutogradMLC, AutogradNestedTensor, AutogradPrivateUse1, AutogradPrivateUse2, AutogradPrivateUse3, Tracer, Autocast, Batched, VmapMode].

CPU: registered at aten\src\ATen\RegisterCPU.cpp:9269 [kernel]
BackendSelect: fallthrough registered at ..\aten\src\ATen\core\BackendSelectFallbackKernel.cpp:3 [backend fallback]
Named: fallthrough registered at ..\aten\src\ATen\core\NamedRegistrations.cpp:11 [kernel]
InplaceOrView: registered at ..\torch\csrc\autograd\generated\InplaceOrViewType_1.cpp:2056 [kernel]
AutogradOther: registered at ..\torch\csrc\autograd\generated\VariableType_4.cpp:8844 [autograd kernel]
AutogradCPU: registered at ..\torch\csrc\autograd\generated\VariableType_4.cpp:8844 [autograd kernel]
AutogradCUDA: registered at ..\torch\csrc\autograd\generated\VariableType_4.cpp:8844 [autograd kernel]
AutogradXLA: registered at ..\torch\csrc\autograd\generated\VariableType_4.cpp:8844 [autograd kernel]
UNKNOWN_TENSOR_TYPE_ID: registered at ..\torch\csrc\autograd\generated\VariableType_4.cpp:8844 [autograd kernel]
AutogradMLC: registered at ..\torch\csrc\autograd\generated\VariableType_4.cpp:8844 [autograd kernel]
AutogradNestedTensor: registered at ..\torch\csrc\autograd\generated\VariableType_4.cpp:8844 [autograd kernel]
AutogradPrivateUse1: registered at ..\torch\csrc\autograd\generated\VariableType_4.cpp:8844 [autograd kernel]
AutogradPrivateUse2: registered at ..\torch\csrc\autograd\generated\VariableType_4.cpp:8844 [autograd kernel]
AutogradPrivateUse3: registered at ..\torch\csrc\autograd\generated\VariableType_4.cpp:8844 [autograd kernel]
Tracer: registered at ..\torch\csrc\autograd\generated\TraceType_4.cpp:9609 [kernel]
Autocast: fallthrough registered at ..\aten\src\ATen\autocast_mode.cpp:250 [backend fallback]
Batched: registered at ..\aten\src\ATen\BatchingRegistrations.cpp:1016 [backend fallback]
VmapMode: fallthrough registered at ..\aten\src\ATen\VmapModeRegistrations.cpp:33 [backend fallback]

The main issue seem to be:

Could not run 'aten::abs.out' with arguments from the 'Vulkan' backend. 

I checked and found not abs.out implementation in 'Vulkan' backend

[skyline75489]

OK I get it what's happening. The Vulkan backend is actually working:

>>> import torch
>>> a = torch.rand(1,3)
>>> a
tensor([[0.1951, 0.5828, 0.5201]])
>>> t = a.to('vulkan')
>>> t.device
device(type='vulkan')

But the missing operators causes the error when using __str__ (or __repr__?), which makes it very difficult to use.