ramjet.losses

Code for custom losses for the ramjet package.

Module Contents

class PerTimeStepBinaryCrossEntropy(positive_weight: float = 1, *args, **kwargs)

Bases: tensorflow.python.keras.losses.LossFunctionWrapper

Computes the cross-entropy loss between true labels and predicted labels for a time series which each time step has a binary label.

__init__(self, positive_weight: float = 1, *args, **kwargs)

Parameters:positive_weight – The weight to give to positive labels in calculating the loss (relative to negative).

static per_time_step_binary_cross_entropy(y_true: tf.Tensor, y_pred: tf.Tensor, positive_weight: float = 1)

Calculates the cross-entropy loss between true labels and predicted labels for a time series which each time step has a binary label.

Parameters:

• y_true – The true label.
• y_pred – The predicted label.
• positive_weight – The weight to give to positive labels in calculating the loss (relative to negative).

Returns:

The resulting cross entropy loss.