Inversion

class cuequivariance_torch.Inversion(

irreps: Irreps,

*,

layout: IrrepsLayout | None = None,

layout_in: IrrepsLayout | None = None,

layout_out: IrrepsLayout | None = None,

device: device | None = None,

math_dtype: dtype | None = None,

use_fallback: bool | None = None,

)

Inversion layer for \(O(3)\) representations.

Parameters:

• irreps (Irreps) – The irreducible representations of the tensor to invert.

• layout (IrrepsLayout, optional) – The memory layout of the tensor, cue.ir_mul is preferred.

• use_fallback (bool, optional) – If None (default), a CUDA kernel will be used if available. If False, a CUDA kernel will be used, and an exception is raised if it’s not available. If True, a PyTorch fallback method is used regardless of CUDA kernel availability.

Forward Pass

forward(x: Tensor) → Tensor

Apply the inversion layer.