# MobileNet

**Repository Path**: zjh56_admin/MobileNet

## Basic Information

- **Project Name**: MobileNet
- **Description**: MobileNet with Re-training/Fine-tuning and Center/Triplet Loss
- **Primary Language**: Unknown
- **License**: MIT
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2020-11-02
- **Last Updated**: 2020-12-19

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

<img src="https://upload.wikimedia.org/wikipedia/commons/thumb/2/2d/Tensorflow_logo.svg/1000px-Tensorflow_logo.svg.png" align="right" width="100"/>

# MobileNet ![You_asked for_this](https://img.shields.io/badge/You_asked-for_this-orange.svg)


> MobileNet on TensorFlow with ability to fine-tune and incorporate center or triplet loss

A tensorflow implementation of Google's [MobileNets](https://arxiv.org/abs/1704.04861) for re-training/fine-tuning on your own custom dataset with the addition of (optional) center loss or triplet loss. Additionally, this repo can be used to re-train Inception network as well with the above added benefits.

## <img src="https://upload.wikimedia.org/wikipedia/commons/thumb/2/2d/Tensorflow_logo.svg/1000px-Tensorflow_logo.svg.png" width="25"/> Tensorflow release
Currently this repo is compatible with Tensorflow 1.3.0.

## <img src="https://image.flaticon.com/icons/svg/149/149366.svg" width="25"/> News
| Date     | Update |
|----------|--------|
| 2017-10-25 | Currently working on triplet loss |
| 2017-10-25 | Added code to support center loss |

## <img src="https://image.flaticon.com/icons/svg/48/48541.svg" width="25"/> Pre-trained Model
Inception_v3 is the most accurate model, but also the slowest. For faster or smaller models, choose a MobileNet with the form `mobilenet_<parameter_size>_<input_size>_[(optional)quantized]`. For example,'mobilenet_1.0_224' will pick a model that is 17 MB in size and takes 224
     pixel input images, while 'mobilenet_0.25_128_quantized' will choose a much
      less accurate, but smaller and faster network that's 920 KB on disk and
      takes 128x128 images.

These models are automatically downloaded for you.

## <img src="https://image.flaticon.com/icons/svg/1/1383.svg" width="25"/> Installation
Details on how to install and re-train on your own dataset can be found on the ![wiki page](https://github.com/ronny3050/MobileNet/wiki/Re-train-MobileNet-on-your-own-dataset). Different parameters that can be tweaked are also given there.

## <img src="https://image.flaticon.com/icons/svg/66/66298.svg" width="25"/> Inspiration
The code is heavily inspired by the Tensorflow's [Retrain Script](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/examples/image_retraining/retrain.py) and [FaceNet](https://github.com/davidsandberg/facenet).