torch.conj — PyTorch 2.7 documentation (original) (raw)

torch.conj(input) → Tensor¶

Returns a view of input with a flipped conjugate bit. If input has a non-complex dtype, this function just returns input.

Note

torch.conj() performs a lazy conjugation, but the actual conjugated tensor can be materialized at any time using torch.resolve_conj().

Warning

In the future, torch.conj() may return a non-writeable view for an input of non-complex dtype. It’s recommended that programs not modify the tensor returned by torch.conj_physical()when input is of non-complex dtype to be compatible with this change.

Parameters

input (Tensor) – the input tensor.

Example:

x = torch.tensor([-1 + 1j, -2 + 2j, 3 - 3j]) x.is_conj() False y = torch.conj(x) y.is_conj() True