odol0503 / caffe2-mobilenet

MobileNets Reimplementation

caffe2-mobilenet

Intro

This is a Caffe2 implementation of Google's MobileNets.For details, please read the original paper:
"MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications" by Andrew G. Howard et. al. 2017
This is refer to shicai/MobileNet-Caffe. My code is not elegant.

Loss and accuracy

I provide a pretrained MobileNet model on cifar10, because of smaller image size than imagenet, I reduce some downsampling operation, and got 89.24% accuracy.

How to Use

python ./mobile_net_trainer.py --train_data [path-to-train-dataset] --test_data [path-to-test-dataset]

Citation

@article{Howard2017mobilenet,
  Author = {Andrew G. Howard, Menglong Zhu, Bo Chen, Dmitry Kalenichenko, Weijun Wang, Tobias Weyand, Marco Andreetto, Hartwig Adam},
  Journal = {arXiv preprint arXiv:1704.04861},
  Title = {MobileNets: Efficient Convolutional Neural Networks for Mobile Vision},
  Year = {2017}
}