02 exercise make_moons, worked out a tanh function but can't get the argument in the model

[RayWagner]

(short comment: the posting removes the indentation)

I got this from the documentation on tanh

def tanh(z):
return (torch.exp(z) - torch.exp(-z)) / (torch.exp(z) + torch.exp(-z))

The original line in the doc was nn.exp but I got numpy array errors then which is a pain at that point in the code. So I tried torch.exp

I put this in a cell.
Then in the model def I try using it.

class MoonModelV1(nn.Module):
def init(self):
super().init()
#out_features = 5 expands to have nore nodes, upscale
self.layer_1 = nn.Linear(in_features=NUM_FEATURES, out_features=10)
self.layer_2 = nn.Linear(in_features=10, out_features=10)
self.layer_3 = nn.Linear(in_features = 10, out_features = NUM_CLASSES)
self.relu = nn.ReLU()
self.tanh = torch.tanh()

def forward(self, x):
    return self.layer_3(self.relu(self.layer_2(self.relu(self.layer_1(x)))))

I get this error
TypeError: tanh() missing 1 required positional arguments: "input"
I get the same error when I replace relu in the forward section too.

I can see it has an input but I don't know what variable that should be or how to provide it when calling the model.
We did some other functions ourselves in the course but I don't remember seeing this as a problem, relu for example.
I get a good result not using tamh anyway, but I do want to know how to write my own functions and be able to use them in a model.