Adjoint

Returns a view of the tensor conjugated and with the last two dimensions transposed.

Abstract Signature:

Adjoint(input: Tensor)

PyTorchJAX (Core)NumPyKerasTensorFlowApple MLXFlax NNXPaxML / Praxis

PyTorch

API: torch.adjoint

Strategy: Direct Mapping

Official Docs ↗

JAX (Core)

API: —

Strategy: Macro 'jnp.conjugate(jnp.swapaxes({input}, -1, -2))'

NumPy

API: —

Strategy: Macro 'np.conjugate(np.swapaxes({input}, -1, -2))'

Keras

API: —

Strategy: Macro 'keras.ops.conj(keras.ops.swapaxes({input}, -1, -2))'

TensorFlow

API: —

Strategy: Macro 'tf.math.conj(tf.linalg.adjoint({input}))'

Apple MLX

API: —

Strategy: Macro 'mx.conjugate(mx.swapaxes({input}, -1, -2))'

Flax NNX

API: —

Strategy: Macro 'jnp.conjugate(jnp.swapaxes({input}, -1, -2))'

PaxML / Praxis

API: —

Strategy: Macro 'jnp.conjugate(jnp.swapaxes({input}, -1, -2))'