Loss

class Loss

Bases: object

Base loss class that all other losses inherit.

Example:

class CustomLoss(Loss):
    def loss(self, predicted: Tensor, actual: Tensor) -> float:
        ... loss function

    def grad(self, predicted: Tensor, actual: Tensor) -> float:
        ... gradient of loss function

__init__()

Initialize self. See help(type(self)) for accurate signature.

Methods

__init__()	Initialize self.
grad(predicted, actual)	Calculates the loss function.
loss(predicted, actual)	Calculates the loss function.

grad(predicted: numpy.ndarray, actual: numpy.ndarray) → float

Calculates the loss function.

Parameters

• predicted (Tensor) – models output

• actual (Tensor) – expected output

Raises

NotImplementedError –

loss(predicted: numpy.ndarray, actual: numpy.ndarray) → float

Calculates the loss function.

Parameters

• predicted (Tensor) – models output

• actual (Tensor) – expected output

Raises

NotImplementedError –