Introduction

This is "Fewer is More: A Deep Graph Metric Learning Perspective Using Fewer Proxies", official pytorch implementation of ProxyGML loss for deep metric learning. The paper is accepted by NeurIPS 2020 (spotlight).

Requirements and Installation

We recommended the following dependencies.

• PyTorch==1.2
• Pillow==5.2.0
• tqdm==4.26.0
• matplotlib==2.2.2
• pandas==0.23.4
• scipy==1.2.1
• scikit-learn==0.20.3
• scikit-image==0.14.2
• h5py==2.9.0

The pipline of ProxyGML

Datasets

1. Download three public benchmarks for deep metric learning

   • CUB-200-2011
   • Cars-196 (Img, Annotation)
   • Stanford Online Products (Link)

2. All datasets are preprocessed as follows ( take CUB200-2011 for eaxample) and stored in .\data

cub200                                         
└───train
|    └───0
|           │   xxx.jpg
|           │   ...
|
|    ...
|
|    └───99
|           │   xxx.jpg
|           │   ...

└───test
|    └───100
|           │   xxx.jpg
|           │   ...
|
|    ...
|
|    └───199
|           │   xxx.jpg
|           │   ...
|    ...

Training Embedding Network

Note that a fine-grained combination of parameter N and parameter r resulted in better overall performance than that described in the paper.

CUB-200-2011

python train.py -b 32 --gpu 2 --dataset cub200 --freeze_BN --epochs 50 --dim 512 --r 0.05 -C 100 --N 12 --weight_lambda 0.3 --centerlr 0.03 --rate 0.1 --new_epoch_to_decay 20 40

Cars-196

python train.py -b 32 --gpu 2 --dataset cars196 --freeze_BN --epochs 50 --dim 512 --r 0.05 -C 98 --N 12 --weight_lambda 0.3 --centerlr 0.03 --rate 0.1 --new_epoch_to_decay 20 40

Stanford Online Products

python train.py -b 32 --gpu 2 --dataset online_products --epochs 50 --dim 512 --r 0.05 -C 11318 --N 1 --weight_lambda 0.0 --centerlr 0.3 --rate 0.1 --new_epoch_to_decay 20 40