Caffe2 is a lightweight, modular, and scalable deep learning framework.

Setup and customize deep learning environment in seconds.

In this repository, I will share some useful notes and references about deploying deep learning-based models in production.

The convertor/conversion of deep learning models for different deep learning frameworks/softwares.

Pretrained EfficientNet, EfficientNet-Lite, MixNet, MobileNetV3 / V2, MNASNet A1 and B1, FBNet, Single-Path NAS

nGraph has moved to OpenVINO

Samples and Tools for Windows ML.

Python library to easily log experiments and parallelize hyperparameter search for neural networks

🔀 Neural Network (NN) Streamer, Stream Processing Paradigm for Neural Network Apps/Devices.

PyTorch implementation of the YOLO (You Only Look Once) v2

Caffe2 on iOS Real-time Demo. Test with Your Own Model and Photos.

Caffe2 implementation of Open Neural Network Exchange (ONNX)

Deep Learning Benchmarking Suite

[EXPERIMENTAL] Demo of using PyTorch 1.0 inside an Android app. Test with your own deep neural network such as ResNet18/SqueezeNet/MobileNet v2 and a phone camera.

Pytorch implementation of the DeepDream computer vision algorithm

PyTorch implementation of the OpenPose

☕️ Caffe2Kit. A simple one step integration of Caffe2 for iOS.

Convert image folder to lmdb, adapted from Efficient-PyTorch

Show how to use Caffe2 in C++ through a simple LeNet sample project

CNN model inference benchmarks for some popular deep learning frameworks