search

Oneflow-Inc / models

Models and examples built with OneFlow
Apache License 2.0
94 stars 37 forks source link

readme

OneFlow-Models

Models and examples implement with OneFlow(version >= 0.5.0).

Introduction

English | 简体中文

OneFlow-Models is an open source repo which contains official implementation of different models built on OneFlow. In each model, we provide at least two scripts train.sh and infer.sh for a quick start. For each model, we provide a detailed README to introduce the usage of this model.

Features

Quick Start

Please check our the following demos for a quick start

Model List

Image Classification - [Lenet](https://github.com/Oneflow-Inc/models/blob/main/Demo/quick_start_demo_lenet/lenet.py) - [Alexnet](https://github.com/Oneflow-Inc/models/tree/main/Vision/classification/image/alexnet) - [VGG16/19](https://github.com/Oneflow-Inc/models/tree/main/Vision/classification/image/vgg) - [Resnet50](https://github.com/Oneflow-Inc/models/tree/main/Vision/classification/image/resnet50) - [InceptionV3](https://github.com/Oneflow-Inc/models/tree/main/Vision/classification/image/inception_v3) - [Densenet](https://github.com/Oneflow-Inc/models/tree/main/Vision/classification/image/densenet) - [Resnext50_32x4d](https://github.com/Oneflow-Inc/models/tree/main/Vision/classification/image/resnext50_32x4d) - [Shufflenetv2](https://github.com/Oneflow-Inc/models/tree/main/Vision/classification/image/shufflenetv2) - [MobilenetV2](https://github.com/Oneflow-Inc/models/tree/main/Vision/classification/image/mobilenetv2) - [mobilenetv3](https://github.com/Oneflow-Inc/models/tree/main/Vision/classification/image/mobilenetv3) - [Ghostnet](https://github.com/Oneflow-Inc/models/tree/main/Vision/classification/image/ghostnet) - [RepVGG](https://github.com/Oneflow-Inc/models/tree/main/Vision/classification/image/repvgg) - [DLA](https://github.com/Oneflow-Inc/models/tree/main/Vision/classification/image/DLA) - [PoseNet](https://github.com/Oneflow-Inc/models/tree/main/Vision/classification/image/poseNet) - [Scnet](https://github.com/Oneflow-Inc/models/tree/main/Vision/classification/image/scnet) - [Mnasnet](https://github.com/Oneflow-Inc/models/tree/main/Vision/classification/image/mnasnet) - [ViT](https://github.com/Oneflow-Inc/models/tree/main/Vision/classification/image/ViT)
Video Classification - [TSN](https://github.com/Oneflow-Inc/models/tree/main/Vision/classification/video/TSN)
Object Detection - [CSRNet](https://github.com/Oneflow-Inc/models/tree/main/Vision/detection/CSRNet)
Semantic Segmentation - [FODDet](https://github.com/Oneflow-Inc/models/tree/main/Vision/segmentation/FODDet) - [FaceSeg](https://github.com/Oneflow-Inc/models/tree/main/Vision/segmentation/FaceSeg) - [U-Net](https://github.com/Oneflow-Inc/models/tree/main/Vision/segmentation/U-Net)
Generative Adversarial Networks - [DCGAN](https://github.com/Oneflow-Inc/models/tree/main/Vision/gan/DCGAN) - [SRGAN](https://github.com/Oneflow-Inc/models/tree/main/Vision/gan/SRGAN) - [Pix2Pix](https://github.com/Oneflow-Inc/models/tree/main/Vision/gan/Pix2Pix) - [CycleGAN](https://github.com/Oneflow-Inc/models/tree/main/Vision/gan/CycleGAN)
Neural Style Transform - [FastNeuralStyle](https://github.com/Oneflow-Inc/models/tree/main/Vision/style_transform/fast_neural_style)
Person Re-identification - [BoT](https://github.com/Oneflow-Inc/models/tree/main/Vision/reid/BoT)
Natural Language Processing - [RNN](https://github.com/Oneflow-Inc/models/tree/main/NLP/rnn) - [Seq2Seq](https://github.com/Oneflow-Inc/models/tree/main/NLP/seq2seq) - [LSTMText](https://github.com/Oneflow-Inc/models/tree/main/NLP/LSTMText) - [TextCNN](https://github.com/Oneflow-Inc/models/tree/main/NLP/TextCNN) - [Transformer](https://github.com/Oneflow-Inc/models/tree/main/NLP/Transformer) - [Bert](https://github.com/Oneflow-Inc/models/tree/main/NLP/bert-oneflow) - [CPT](https://github.com/Oneflow-Inc/models/tree/main/NLP/CPT) - [Roberta](https://github.com/Oneflow-Inc/models/tree/main/NLP/roberta) - [MoE](https://github.com/Oneflow-Inc/models/tree/main/NLP/MoE)
Audio - [SincNet](https://github.com/Oneflow-Inc/models/tree/main/Audio/SincNet) - [Wav2Letter](https://github.com/Oneflow-Inc/models/tree/main/Audio/Wav2Letter) - [AM_MobileNet1D](https://github.com/Oneflow-Inc/models/tree/main/Audio/AM-MobileNet1D) - [Speech-Emotion-Analyer](https://github.com/Oneflow-Inc/models/tree/main/Audio/Speech-Emotion-Analyzer) - [Speech-Transformer](https://github.com/Oneflow-Inc/models/tree/main/Audio/Speech-Transformer) - [CycleGAN-VC2](https://github.com/Oneflow-Inc/models/tree/main/Audio/CycleGAN-VC2) - [MaskCycleGAN-VC](https://github.com/Oneflow-Inc/models/tree/main/Audio/MaskCycleGAN-VC) - [StarGAN-VC](https://github.com/Oneflow-Inc/models/tree/main/Audio/StarGAN-VC) - [Adaptive_Voice_Conversion](https://github.com/Oneflow-Inc/models/tree/main/Audio/Adaptive_Voice_Conversion)
Deep Reinforcement Learning - [FlappyBird](https://github.com/Oneflow-Inc/models/tree/main/DeepReinforcementLearning/FlappyBird)
Quantization Aware Training - [Quantization](https://github.com/Oneflow-Inc/models/tree/main/Quantization)
Knowledge Distillation - [KnowledgeDistillation](https://github.com/Oneflow-Inc/models/tree/main/KnowledgeDistillation/KnowledgeDistillation)

Installation and Environment setup

Install Oneflow

https://github.com/Oneflow-Inc/oneflow#install-with-pip-package

Build custom ops from source

In the root directory, run:

mkdir build
cd build
cmake ..
make -j$(nrpoc)

Example of using ops:

from ops import RoIAlign
pooler = RoIAlign(output_size=(14, 14), spatial_scale=2.0, sampling_ratio=2)