|
Application
|
Models
|
|
CV
|
Resnet50, Shufflenet, Squeezenet, Densenet, Efficientnet, Mobilenet_v1, Mobilenet_v2, Mobilenet_v3,
BiRealNet,
ReActNet,
Ghostnet,
unet,
LCNet, Pointnet,
hair-segmentation,
duc,
fcn,
retinanet,
SSD,
Faster-RCNN,
Mask-RCNN,
Yolov2,
Yolov3,
Yolov4,
Yolov5,
ViT, TNT, RepVGG, VitAE, CMT, EfficientFormer ...
|
|
NLP
|
Bert, Albert, Tinybert, Neural Machine Translation, Text To Speech(Tactron,Tactron2,FastSpeech+hifigan,melgan), Automatic Speech Recognition, DFSMN, Conformer,
Tdnn,
FRILL,
T5,
GPT-2,
Roberta, Wenet ...
|
|
Recommendation
|
NFM, AFM, ONN, wide&deep, DeepFM, MMOE
|
|
More DL Tasks
|
...
|
More models than these mentioned above are supported, users are encouraged to further explore.
# On-Device Training
---
On-Device Training has come, it's a beta vesion which supports [Lenet](./training/demos/lenet_demo/), [Mobilenet_v1](./training/demos/mobilenet_v1_demo) and [Resnet18](./training/demos/resnet18_demo) for training on the embedded devices and servers. Want more details of on-device training in bolt? Get with the official training [tutorial](./training/TUTORIAL.md).
# Documentations
---
Everything you want to know about bolt is recorded in the detailed documentations stored in [docs](docs).
- [How to install bolt with different compilers?](docs/INSTALL.md).
- [How to use bolt to inference your ML models?](docs/USER_HANDBOOK.md)
- [How to deploy Pytorch model?](docs/DEPLOYMENT_GUIDE_PYTORCH_CN.md)
- [How to deploy Tensorflow model?](docs/DEPLOYMENT_GUIDE_TENSORFLOW_CN.md)
- [How to deploy Onnx model?](docs/DEPLOYMENT_GUIDE_ONNX_CN.md)
- [How to deploy Tflite model?](docs/DEPLOYMENT_GUIDE_TFLITE_CN.md)
- [How to deploy Caffe model?](docs/DEPLOYMENT_GUIDE_CAFFE_CN.md)
- [How to develop bolt to customize more models?](docs/DEVELOPER.md)
- [Operators documentation](docs/OPERATORS.md)
- [Benchmark results on some universal models.](docs/BENCHMARK.md)
- [How to visualise/protect bolt model?](docs/USER_HANDBOOK.md#model-visualization)
- [How to build demo/example with kit?](docs/KIT.md)
- [Frequently Asked Questions(FAQ)](docs/FAQ.md)
# Articles
---
- [深度学习加速库Bolt领跑端侧AI](https://zhuanlan.zhihu.com/p/317111024)
- [为什么 Bolt 这么快：矩阵向量乘的实现](https://zhuanlan.zhihu.com/p/97928435)
- [深入硬件特性加速TinyBert，首次实现手机上Bert 6ms推理](https://zhuanlan.zhihu.com/p/158620259)
- [Bolt GPU性能优化，让上帝帮忙掷骰子](https://zhuanlan.zhihu.com/p/336218879)
- [Bolt助力HMS机器翻译，自然语言处理又下一城](https://zhuanlan.zhihu.com/p/337887620)
- [ARM CPU 1-bit推理，走向极致的道路](https://zhuanlan.zhihu.com/p/158161592)
- [基于深度学习加速库Bolt的声音克隆技术(Voice Cloning)](https://zhuanlan.zhihu.com/p/498919929)
# 教程
---
- 图像分类: [Android Demo](https://zhuanlan.zhihu.com/p/359264809), [iOS Demo](https://zhuanlan.zhihu.com/p/359259766)
- 图像增强: [Android Deme](./docs/DIT.md#camera-enlarge), [iOS Demo](./docs/DIT.md#camera-enlarge)
- 情感分类: [Android Demo](https://zhuanlan.zhihu.com/p/414971037)
- 中文语音识别: [Android Demo](https://zhuanlan.zhihu.com/p/414978782), [iOS Demo](https://zhuanlan.zhihu.com/p/414981121)
- 人脸检测: [Android Demo](https://zhuanlan.zhihu.com/p/414975102), [iOS Demo](https://zhuanlan.zhihu.com/p/414971375)
# Acknowledgement
---
Bolt refers to the following projects: [caffe](https://github.com/BVLC/caffe), [onnx](https://github.com/onnx/onnx), [tensorflow](https://github.com/tensorflow/tensorflow), [ncnn](https://github.com/Tencent/ncnn), [mnn](https://github.com/alibaba/MNN), [dabnn](https://github.com/JDAI-CV/dabnn).
# License
---
The MIT License(MIT)
| 
| 
|
| 
| 
| 
|
# Verified Networks
---
Bolt has shown its high performance in the inference of common CV, NLP and Recommendation neural networks. Some of the representative networks that we have verified are listed below. You can find detailed benchmark information in [docs/BENCHMARK.md](docs/BENCHMARK.md).