quantax.model.ResProd#

quantax.model.ResProd(nblocks: int, channels: int, kernel_size: int, final_actfn: ~typing.Callable, dtype: ~numpy.dtype = <class 'jax.numpy.float32'>)#

The convolutional residual network with a product in the end. This network still requires further tests.

Parameters:
  • nblocks – The number of residual blocks. Each block contains two convolutional layers.

  • channels – The number of channels. Each layer has the same amount of channels.

  • kernel_size – The kernel size. Each layer has the same kernel size.

  • final_actfn – The activation function in the last layer.

  • dtype – The data type of the parameters.