def lossGraph(self):
'''
Loss Graph for given Bonsai Obj
'''
self.regLoss = 0.5 * (self.lZ * tf.square(tf.norm(self.bonsaiObj.Z)) +
self.lW * tf.square(tf.norm(self.bonsaiObj.W)) +
self.lV * tf.square(tf.norm(self.bonsaiObj.V)) +
self.lT * tf.square(tf.norm(self.bonsaiObj.T)))
if (self.bonsaiObj.numClasses > 2):
if self.useMCHLoss is True:
self.batch_th = tf.placeholder(tf.int64, name='batch_th')
self.marginLoss = utils.multiClassHingeLoss(
tf.transpose(self.score), self.Y,
self.batch_th)
else:
self.marginLoss = utils.crossEntropyLoss(
tf.transpose(self.score), self.Y)
self.loss = self.marginLoss + self.regLoss
else:
self.marginLoss = tf.reduce_mean(tf.nn.relu(
1.0 - (2 * self.Y - 1) * tf.transpose(self.score)))
self.loss = self.marginLoss + self.regLoss
return self.loss, self.marginLoss, self.regLoss
def lossGraph(self):
'''
Loss Graph for given Bonsai Obj
'''
self.regLoss = 0.5 * (self.lZ * tf.square(tf.norm(self.bonsaiObj.Z)) +
self.lW * tf.square(tf.norm(self.bonsaiObj.W)) +
self.lV * tf.square(tf.norm(self.bonsaiObj.V)) +
self.lT * tf.square(tf.norm(self.bonsaiObj.T)))
# Loss functions for classification.
if (self.bonsaiObj.isRegression is False):
if (self.bonsaiObj.numClasses > 2):
if self.useMCHLoss is True:
self.batch_th = tf.placeholder(tf.int64, name='batch_th')
self.marginLoss = utils.multiClassHingeLoss(
tf.transpose(self.score), self.Y,
self.batch_th)
else:
self.marginLoss = utils.crossEntropyLoss(
tf.transpose(self.score), self.Y)
self.loss = self.marginLoss + self.regLoss
else:
self.marginLoss = tf.reduce_mean(tf.nn.relu(
1.0 - (2 * self.Y - 1) * tf.transpose(self.score)))
self.loss = self.marginLoss + self.regLoss
# Loss functions for regression.
elif (self.bonsaiObj.isRegression is True):
if(self.regressionLoss == 'huber'):
# Use of Huber Loss , because it is more robust to outliers.
self.marginLoss = tf.losses.huber_loss(
self.Y, tf.transpose(self.score))
self.loss = self.marginLoss + self.regLoss
elif (self.regressionLoss == 'l2'):
# L2 loss function.
self.marginLoss = tf.nn.l2_loss(
self.Y - tf.transpose(self.score))
self.loss = self.marginLoss + self.regLoss
return self.loss, self.marginLoss, self.regLoss