Last active
April 24, 2019 12:07
-
-
Save amineHY/ffeb2d68b20f4255b14cfe3ad09cadff to your computer and use it in GitHub Desktop.
This function implement the computation of the triplet loss function for the face recognition application
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
def triplet_loss_function(im_anchor, im_positive, im_negative, alpha = 0.2): | |
""" | |
Implementation of the triplet loss function | |
Source: https://arxiv.org/pdf/1503.03832.pdf | |
Arguments: | |
y_pred -- python list containing three objects: | |
anchor -- the encodings for the anchor images, of shape (None, 128) | |
positive -- the encodings for the positive images, of shape (None, 128) | |
negative -- the encodings for the negative images, of shape (None, 128) | |
Returns: | |
loss -- real number, value of the loss | |
""" | |
# Compute the (encoding) distance between the anchor and the positive, then negative images | |
pos_dist = tf.reduce_sum(tf.square(tf.subtract(im_anchor, im_positive)), axis=-1) | |
neg_dist = tf.reduce_sum(tf.square(tf.subtract(im_anchor, im_negative)), axis=-1) | |
basic_loss = tf.add(tf.subtract(pos_dist, neg_dist), alpha) | |
return tf.reduce_sum(tf.maximum(basic_loss, 0.0)) | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment