TorchScript Unsupported Pytorch Constructs¶
Torch and Tensor Unsupported Attributes¶
TorchScript supports most methods defined on
torch.Tensor, but we do not have full coverage.
Here are specific known ops and categories of ops which have diverging behavior between
Python and TorchScript. If you encounter something else that is not supported please
file a GitHub issue. Deprecated ops are not listed below.
Unsupported Tensor Methods¶
Functions Not Correctly Bound on Torch¶
The following functions will fail if used in TorchScript, either because they are not bound on torch or because Python expects a different schema than TorchScript.
Ops With Divergent Schemas Between Torch & Python¶
The following categories of ops have divergent schemas:
Functions which construct tensors from non-tensor inputs do not support the requires_grad argument, except for torch.tensor. This covers the following ops:
The following functions require dtype, layout, device as parameters in TorchScript, but these parameters are optional in Python.
PyTorch Unsupported Modules and Classes¶
TorchScript cannot currently compile a number of other commonly used PyTorch
constructs. Below are listed the modules that TorchScript does not support, and
an incomplete list of PyTorch classes that are not supported. For unsupported modules
we suggest using