TEE-AI / Sai
Licence: apache-2.0
SDK for TEE AI Stick (includes model training script, inference library, examples)
Stars: ✭ 28
Programming Languages
python
139335 projects - #7 most used programming language
Projects that are alternatives of or similar to Sai
Jacinto Ai Devkit
Training & Quantization of embedded friendly Deep Learning / Machine Learning / Computer Vision models
Stars: ✭ 49 (+75%)
Mutual labels: cnn, quantization
Awesome Edge Machine Learning
A curated list of awesome edge machine learning resources, including research papers, inference engines, challenges, books, meetups and others.
Stars: ✭ 139 (+396.43%)
Mutual labels: quantization, iot
Lq Nets
LQ-Nets: Learned Quantization for Highly Accurate and Compact Deep Neural Networks
Stars: ✭ 195 (+596.43%)
Mutual labels: cnn, quantization
Tf2
An Open Source Deep Learning Inference Engine Based on FPGA
Stars: ✭ 113 (+303.57%)
Mutual labels: cnn, quantization
Model Quantization
Collections of model quantization algorithms
Stars: ✭ 118 (+321.43%)
Mutual labels: cnn, quantization
Awareness
The new architecture of co-computation for data processing and machine learning.
Stars: ✭ 11 (-60.71%)
Mutual labels: iot
Channelslightscontrol
Demo app with Django Channels to control Lights over websockets. Made for PyStPete meetup(https://www.meetup.com/Saint-Petersburg-Python-Meetup/).
Stars: ✭ 14 (-50%)
Mutual labels: iot
Training extensions
Trainable models and NN optimization tools
Stars: ✭ 857 (+2960.71%)
Mutual labels: quantization
Attention Ocr
A Tensorflow model for text recognition (CNN + seq2seq with visual attention) available as a Python package and compatible with Google Cloud ML Engine.
Stars: ✭ 844 (+2914.29%)
Mutual labels: cnn
Esp8266 thing dev
An all-in-one development board for the ESP8266 including an FTDI FT231X for USB programming.
Stars: ✭ 11 (-60.71%)
Mutual labels: iot
Wifi Connect
Easy WiFi setup for Linux devices from your mobile phone or laptop
Stars: ✭ 883 (+3053.57%)
Mutual labels: iot
Easyntpclient
Library to read time from Network Time Protocol (NTP) servers.
Stars: ✭ 20 (-28.57%)
Mutual labels: iot
Worldpay Within Sdk
Worldpay Within SDK to allow payments within IoT. Written in Go.
Stars: ✭ 12 (-57.14%)
Mutual labels: iot
Neural Image Captioning
Implementation of Neural Image Captioning model using Keras with Theano backend
Stars: ✭ 12 (-57.14%)
Mutual labels: cnn
TEE Simple AI (SAI) User Manual
SAI是基于PyTorch/Caffe的卷积神经网络模型训练，转换，部署工具——可训练得到定点化模型（1bit或者3bit），该模型可在TEE.COM的算力棒产品上运行。支持Windows，Linux等主流平台。
TODO list:
- [x] PyTorch quantization model training for classification & conversion
- [x] Caffe quantization model training for classification & conversion
- [x] Windows/Linux classify inference library and c++ examples
- [x] Python examples
- [ ] ARM-Linux/Android inference library and c++ examples
- [x] Caffe Faster RCNN model training & conversion & inference library
- [ ] Pytorch Faster RCNN model training & conversion & inference library
- [ ] Model Zoo
- [ ] Benchmark
算力棒运行环境
系统 (Win10/Ubuntu 16.04)
安装算力棒驱动
Windows 10系统无需安装驱动，Linux系统请参考以下命令为算力棒安装驱动
$ sudo cp api/lib/linux/50-emmc.rules /etc/udev/rules.d/
模型训练(SAI_ROOT/train)
针对不同的开发者，我们提供了两种训练框架(Pytorch与Caffe)的模型训练与转换，大家可以根据个人喜好选择使用。
推断部署(SAI_ROOT/api)
得到可以在算力棒上运行的模型后，我们可以通过调用SAI的API，在终端设备上很方便的部署上你的模型。
目前API里包括用于图片分类的TEEClassifier library和用于目标检测的基于Faster RCNN的TEEDetectorFRCNN library，支持windows/linux/arm-linux三个平台，后续会增加android/ios等平台支持。
在SAI_ROOT/example下提供了各个平台的c++/python示例工程，展示了API的调用方法。
Issue: Windows版本的library依赖于libeay32.dll，请确定系统路径里存在该dll文件,若没有的话，可直接使用api目录下的libeay32.dll。
交流群
QQ群: 781349981
Note that the project description data, including the texts, logos, images, and/or trademarks,
for each open source project belongs to its rightful owner.
If you wish to add or remove any projects, please contact us at [email protected].