losses.py

import tensorflow as tf
from keras import backend
from keras.losses import LossFunctionWrapper
from keras.utils import losses_utils
from tensorflow.python.util.tf_export import keras_export


@keras_export('keras.losses.log_cosh', 'keras.losses.logcosh',
              'keras.metrics.log_cosh', 'keras.metrics.logcosh')
@tf.__internal__.dispatch.add_dispatch_support
def log_cosh(y_true, y_pred):
    y_pred = tf.convert_to_tensor(y_pred)
    y_true = tf.cast(y_true, y_pred.dtype)

    def _log_cosh(x):
        return x + tf.math.softplus(-2. * x) - tf.cast(tf.math.log(2.), x.dtype)

    return backend.mean(_log_cosh(y_pred - y_true), axis=-1)


class LogCosh(LossFunctionWrapper):
    def __init__(self, reduction=losses_utils.ReductionV2.AUTO, name='log_cosh'):
        super().__init__(log_cosh, name=name, reduction=reduction)


@keras_export('keras.losses.bcesque', 'keras.losses.bcesque',
              'keras.metrics.bcesque', 'keras.metrics.bcesque')
@tf.__internal__.dispatch.add_dispatch_support
def bcesque(y_true, y_pred):
    y_pred = tf.convert_to_tensor(y_pred)
    y_true = tf.cast(y_true, y_pred.dtype)
    
    y_label = tf.clip_by_value(y_true, -2., 0.) * 0.5 + 1.
    
    return -backend.mean(y_label * y_pred - tf.math.softplus(y_pred), axis=-1)


class Bcesque(LossFunctionWrapper):
    def __init__(self, reduction=losses_utils.ReductionV2.AUTO, name='bcesque'):
        super().__init__(bcesque, name=name, reduction=reduction)