This repository is an implementation of Deep Metric Learning via Facility Location on tensorflow. We build this on Cifar100 and Densenet-40. This paper is available here. For the loss layer implementation, look at here. For the Densenet implementation, look at here.

@inproceedings{songCVPR17,
  Author    = {Hyun Oh Song and Stefanie Jegelka and Vivek Rathod and Kevin Murphy},
  Title     = {Deep Metric Learning via Facility Location},
  Booktitle = {Computer Vision and Pattern Recognition (CVPR)},
  Year      = {2017}
}

1. Install prerequsites for tensorflow (see: tensorflow-gpu installation instructions).
2. Run pip install -r requirements.txt get required support.

1. Modify metric_learning_densenet.py for training-params and densenet-params. We pick Cifar100 as our training data, because it's tiny, save GPU-memory (when batch size 64, it cost about 4.6G GPU-Memory) and good for doing research.
2. Run python metric_learning_densenet.py, the data_provider with automaticlly handle data download and process. After that, start Densenet-Cluster-loss training.
3. Download Downsampled Imagenet with size 32x32 from here. Modify metric_learning_densenet.py train on Imagenet.

1. Modify metric_learning_densenet.py extract feature embeddings on cifar test set, the embeddings is saved with .npy format used for evaluation process.

1. Run python visualization/tsne.py can plot and save the cluster result on Cifar database.

• Densenet tensorflow training code
• Deep metric learning cluster loss code
• Evaludation
  • NMI, Recall@K code
  • feature extraction code
  • feature visulization code (tSNE)
• Dataset support
  • cifar-10
  • cifar-100
  • imagenet-32x32