Source code for radio.models.keras.losses

""" Contains losses used in keras models. """
from keras import backend as K


def dice_loss(y_true, y_pred, smooth=1e-6):
    """ Loss function base on dice coefficient.

    Parameters
    ----------
    y_true : keras tensor
        tensor containing target mask.
    y_pred : keras tensor
        tensor containing predicted mask.
    smooth : float
        small real value used for avoiding division by zero error.

    Returns
    -------
    keras tensor
        tensor containing dice loss.
    """
    y_true_f = K.flatten(y_true)
    y_pred_f = K.flatten(y_pred)
    intersection = K.sum(y_true_f * y_pred_f)
    answer = (2. * intersection + smooth) / (K.sum(y_true_f) + K.sum(y_pred_f) + smooth)
    return -answer


def tversky_loss(y_true, y_pred, alpha=0.3, beta=0.7, smooth=1e-10):
    """ Tversky loss function.

    Parameters
    ----------
    y_true : keras tensor
        tensor containing target mask.
    y_pred : keras tensor
        tensor containing predicted mask.
    alpha : float
        real value, weight of '0' class.
    beta : float
        real value, weight of '1' class.
    smooth : float
        small real value used for avoiding division by zero error.

    Returns
    -------
    keras tensor
        tensor containing tversky loss.
    """
    y_true = K.flatten(y_true)
    y_pred = K.flatten(y_pred)
    truepos = K.sum(y_true * y_pred)
    fp_and_fn = alpha * K.sum(y_pred * (1 - y_true)) + beta * K.sum((1 - y_pred) * y_true)
    answer = (truepos + smooth) / ((truepos + smooth) + fp_and_fn)
    return -answer


def jaccard_coef_logloss(y_true, y_pred, smooth=1e-10):
    """ Loss function based on jaccard coefficient.

    Parameters
    ----------
    y_true : keras tensor
        tensor containing target mask.
    y_pred : keras tensor
        tensor containing predicted mask.
    smooth : float
        small real value used for avoiding division by zero error.

    Returns
    -------
    keras tensor
        tensor containing negative logarithm of jaccard coefficient.
    """
    y_true = K.flatten(y_true)
    y_pred = K.flatten(y_pred)
    truepos = K.sum(y_true * y_pred)
    falsepos = K.sum(y_pred) - truepos
    falseneg = K.sum(y_true) - truepos
    jaccard = (truepos + smooth) / (smooth + truepos + falseneg + falsepos)
    return -K.log(jaccard + smooth)