conj

Return conjugated tensor with flipped index directions.

conj

conj(tensor: Tensor) -> Tensor

Return a new tensor with conjugated data and flipped index directions.

Parameters:

Name	Type	Description	Default
tensor	Tensor	The input tensor to conjugate.	required

Returns:

Type	Description
Tensor	A new tensor instance with: Conjugated dense blocks (if dtype is complex) All index directions flipped Intertwiners (if present) updated with flipped edge directions and FS phase absorbed into weights (±1, SU(2) only) All other attributes preserved

Notes

This functional version creates a fully independent tensor by cloning all data blocks. For an efficient version that shares underlying tensors, use tensor.conj(in_place=False).

Description

Returns a fully independent tensor with:

• Conjugated block data (for complex dtypes)
• All index directions flipped (OUT ↔ IN)

All data blocks are deep-copied, so the result is completely isolated from the original. For a memory-efficient version that shares storage with the original, use Tensor.conj() (default in_place=False). To modify the tensor in place, use Tensor.conj(in_place=True).

See Also

• Tensor.conj: Method form (shared-storage default, or in-place)
• transpose: Transpose axes
• Examples: Manipulation

Notes

For real dtypes, only directions are flipped. For complex dtypes, data is conjugated and directions flipped.