Deep Metric Learning

Implementation of several methods for deep metric learning in Chainer v4.2.0.

• Proxy-NCA: No Fuss Distance Metric Learning using Proxies
• N-pair loss: Improved Deep Metric Learning with Multi-class N-pair Loss Objective
• Lifted structure: Deep Metric Learning via Lifted Structured Feature Embedding
• Clustering loss: Deep Metric Learning via Facility Location
• Angular loss: Deep Metric Learning with Angular Loss

Dependencies

pip install cupy==4.2.0 chainer==4.2.0 fuel==0.2.0 tqdm