Gilt ♥ Open Source / Deep Learning at GILT

Date: January 2, 2017 Published by



  • Using a siamese network and training the network using a Hinge loss function, the network learns to detect dress similarities.
  • This process is called neural network training and most times, involves different forms of a base algorithm called backpropagation.
  • On top of that, we are also developing systems to detect dress similarity which can be useful for product recommendations.
  • It’s important to note here that images in either the test or the training set always remain in test or in train.
  • We are also investing efforts trying to archive similar results using TensorFlow and GoogleNet v3.


http://tech.gilt.com/machine/learning,/deep/learning/2016/12/22/deep-learning-at-gilt

email