pihnn.utils.MSE

pihnn.utils.MSE(value, true_value=None)

Mean squared error (MSE). Equivalent to torch.nn.MSELoss() except it takes into account empty inputs.

Parameters:

• value (torch.tensor) – Input to evaluate.

• true_value – Reference value to compare the input with. If unassigned, it’s by default zero.

Returns:

mse (float) - MSE error.