reparameters vs no reparameters

def sample(self, return_pretanh_value=False):
    """
    Gradients will and should *not* pass through this operation.

    See https://github.com/pytorch/pytorch/issues/4620 for discussion.
    """
    z = self.normal.sample().detach()

    if return_pretanh_value:
        return torch.tanh(z), z
    else:
        return torch.tanh(z)

def rsample(self, return_pretanh_value=False):
    """
    Sampling in the reparameterization case.
    """
    z = (
        self.normal_mean +
        self.normal_std *
        Normal(
            torch.zeros(self.normal_mean.size(), device=self.normal_mean.device),
            torch.ones(self.normal_std.size(), device=self.normal_mean.device)
        ).sample()
    )
    z.requires_grad_()

    if return_pretanh_value:
        return torch.tanh(z), z
    else:
        return torch.tanh(z)


from : 
offlinerl/neorl