This repository contains an op-for-op PyTorch reimplementation of MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications .
Contains MNIST, CIFAR10&CIFAR100, TinyImageNet_200, MiniImageNet_1K, ImageNet_1K, Caltech101&Caltech256 and more etc.
Please refer to README.md in the data directory for the method of making a dataset.
Both training and testing only need to modify the config.py file.
- line 29:
model_arch_namechange tomobilenet_v1. - line 31:
model_mean_parameterschange to[0.485, 0.456, 0.406]. - line 32:
model_std_parameterschange to[0.229, 0.224, 0.225]. - line 34:
model_num_classeschange to1000. - line 36:
modechange totest. - line 89:
model_weights_pathchange to./results/pretrained_models/MobileNetV1-ImageNet_1K.pth.tar.
python3 test.py- line 29:
model_arch_namechange tomobilenet_v1. - line 31:
model_mean_parameterschange to[0.485, 0.456, 0.406]. - line 32:
model_std_parameterschange to[0.229, 0.224, 0.225]. - line 34:
model_num_classeschange to1000. - line 36:
modechange totrain. - line 50:
pretrained_model_weights_pathchange to./results/pretrained_models/MobileNetV1-ImageNet_1K.pth.tar.
python3 train.py- line 29:
model_arch_namechange tomobilenet_v1. - line 31:
model_mean_parameterschange to[0.485, 0.456, 0.406]. - line 32:
model_std_parameterschange to[0.229, 0.224, 0.225]. - line 34:
model_num_classeschange to1000. - line 36:
modechange totrain. - line 53:
resumechange to./samples/mobilenet_v1-ImageNet_1K/epoch_xxx.pth.tar.
python3 train.pySource of original paper results: https://arxiv.org/pdf/1704.04861v1.pdf)
In the following table, the top-x error value in () indicates the result of the project, and - indicates no test.
| Model | Dataset | Top-1 error (val) | Top-5 error (val) |
|---|---|---|---|
| mobilenet_v1 | ImageNet_1K | 29.4%(-) | -(-) |
# Download `mobilenet_v1.pth.tar` weights to `./results/pretrained_models`
# More detail see `README.md<Download weights>`
python3 ./inference.py Input:
Output:
Build `mobilenet_v1` model successfully.
Load `mobilenet_v1` model weights `/MobileNetV1-PyTorch/results/pretrained_models/mobilenet_v1.pth.tar` successfully.
tench, Tinca tinca (81.23%)
barracouta, snoek (1.41%)
armadillo (0.80%)
dugong, Dugong dugon (0.11%)
rock beauty, Holocanthus tricolor (0.11%)
If you find a bug, create a GitHub issue, or even better, submit a pull request. Similarly, if you have questions, simply post them as GitHub issues.
I look forward to seeing what the community does with these models!
Andrew G. Howard, Menglong Zhu, Bo Chen, Dmitry Kalenichenko, Weijun Wang, Tobias Weyand, Marco Andreetto, Hartwig Adam
We present a class of efficient models called MobileNets for mobile and embedded vision applications. MobileNets are based on a streamlined architecture that uses depth-wise separable convolutions to build light weight deep neural networks. We introduce two simple global hyper-parameters that efficiently trade off between latency and accuracy. These hyper-parameters allow the model builder to choose the right sized model for their application based on the constraints of the problem. We present extensive experiments on resource and accuracy tradeoffs and show strong performance compared to other popular models on ImageNet classification. We then demonstrate the effectiveness of MobileNets across a wide range of applications and use cases including object detection, finegrain classification, face attributes and large scale geo-localization.
@article{howard2017mobilenets,
title={Mobilenets: Efficient convolutional neural networks for mobile vision applications},
author={Howard, Andrew G and Zhu, Menglong and Chen, Bo and Kalenichenko, Dmitry and Wang, Weijun and Weyand, Tobias and Andreetto, Marco and Adam, Hartwig},
journal={arXiv preprint arXiv:1704.04861},
year={2017}
}