insightface/recognition/partial_fc/pytorch

Parital FC

TODO

• No BUG Sampling
• Pytorch Experiments (Glint360k, 1.0/0.1)
• Mixed precision training
• Pipeline Parallel
• Checkpoint
• Docker
• A Wonderful Documents

Results

We employ ResNet100 as the backbone.

1. IJB-C results

Datasets	1e-05	1e-04	1e-03	1e-02	1e-01
Glint360K	95.92	97.30	98.13	98.78	99.28
MS1MV2	94.22	96.27	97.61	98.34	99.08

2. IFRT results

TODO

Training Speed Benchmark

1. Train MS1MV2

Employ ResNet100 as the backbone.

GPU	FP16	BatchSize / it	Throughput img / sec	Time / hours
8 * Tesla V100-SXM2-32GB	False	64	1658	15
8 * Tesla V100-SXM2-32GB	True	64	2243	12
8 * Tesla V100-SXM2-32GB	False	128	1800	14
8 * Tesla V100-SXM2-32GB	True	128	3337	7
8 * RTX2080Ti	False		1200
8 * RTX2080Ti

Employ ResNet50 as the backbone.

GPU	FP16	BatchSize / it	Throughput img / sec	Time / hours
8 * Tesla V100-SXM2-32GB	False	64	2745	9
8 * Tesla V100-SXM2-32GB	True	64	3770	7
8 * Tesla V100-SXM2-32GB	False	128	2833	9
8 * Tesla V100-SXM2-32GB	True	128	5102	5

2. Train millions classes

TODO

How to run

cuda=10.1
pytorch==1.6.0
pip install -r requirement.txt

bash run.sh

使用 bash run.sh 这个命令运行。

Citation

If you find Partial-FC or Glint360K useful in your research, please consider to cite the following related paper:

Partial FC

@inproceedings{an2020partical_fc,
  title={Partial FC: Training 10 Million Identities on a Single Machine},
  author={An, Xiang and Zhu, Xuhan and Xiao, Yang and Wu, Lan and Zhang, Ming and Gao, Yuan and Qin, Bin and
  Zhang, Debing and Fu Ying},
  booktitle={Arxiv 2010.05222},
  year={2020}
}