losses

Classes

class BinaryCrossentropy : A criterion to compute the binary cross entropy with contiguous targets.

class CategoricalCrossentropy : A criterion to compute the categorical cross entropy with contiguous targets.

class Loss : The base class for loss criterion.

class MeanAbsoluteError : A criterion to compute the reduced element-wise absolute value difference.

class MeanSquaredError : A criterion to compute the reduced element-wise squared error.

class SparseCategoricalCrossentropy : A criterion to compute the categorical cross entropy with sparse labels.

Functions

binary_crossentropy(…) : Compute the binary cross entropy with contiguous targets.

categorical_crossentropy(…) : Compute the categorical cross entropy with contiguous targets.

get(…) : Return the loss callable by identifier.

mean_absolute_error(…) : Compute the reduced element-wise absolute value difference.

mean_squared_error(…) : Compute the reduced element-wise squared error.

sparse_categorical_crossentropy(…) : Compute the categorical cross entropy with sparse labels.