I need to implement custom loss function and I saw following tutorial.

```
class LinearFunction(Function):
@staticmethod
def forward(ctx, input, weight, bias=None):
ctx.save_for_backward(input, weight, bias)
output = input.mm(weight.t())
if bias is not None:
output += bias.unsqueeze(0).expand_as(output)
return output
@staticmethod
def backward(ctx, grad_output):
input, weight, bias = ctx.saved_tensors
grad_input = grad_weight = grad_bias = None
if ctx.needs_input_grad[0]:
grad_input = grad_output.mm(weight)
if ctx.needs_input_grad[1]:
grad_weight = grad_output.t().mm(input)
if bias is not None and ctx.needs_input_grad[2]:
grad_bias = grad_output.sum(0).squeeze(0)
return grad_input, grad_weight, grad_bias
```

But I don’t know how to use above guide for my situation.

I have ground truth data, but it’s not a label data.

I perform math formulas with GT data through all image pixels and I obtain one float number.

And I want to update weights in CNN as above float number loss value minimizes.

How to implement forward() and backward() in this case?