v0.4.3

einx v0.4.3

Fixed

• Annotate tensor parameters with typing.Any instead of typing.TypeVar.

  Previously, a typing.TypeVar was used for all input and output tensors of an operation. However, this typing constraint
  does not always hold (e.g., operations accept mixed-type input tensors, Python scalars and tensor factories, and output tensor types depend on dynamic factors such as the selected backend). Tensor type annotations now use typing.Any to reflect the actual behavior.

• Fix typing to also work with ty (#30).

• Add missing py.typed marker to comply with PEP 561 (#29).