quantax.model.SingleConv

quantax.model.SingleConv(channels: int, actfn: ~typing.Callable, use_bias: bool = True, holomorphic: bool = False, dtype: ~numpy.dtype = <class 'jax.numpy.float32'>)

Network with one convolutional layer \(\psi(s) = \prod f(\mathrm{Conv}(s))\).

Parameters:

• features – The number of channels in the convolutional network.

• actfn – The activation function applied after the convolutional layer.

• use_bias – Whether to add on a bias in the convolution.

• holomorphic – Whether the whole network is complex holomorphic.

• dtype – The data type of the parameters.