【资源】AutoML 与轻量模型大列表
分享一份高质量(最新的)AutoML工作和轻量级模型的列表，包括神经结构搜索，轻量级结构，模型压缩和加速，超参数优化，自动特征工程的论文、项目、博客等资源。欢迎star、补充和提出建议。
- 作者：guan-yuan
- 项目地址：awesome-AutoML-and-Lightweight-Models
更多Awsome Github资源请关注：【Awsome】GitHub 资源汇总
1.) 神经结构搜索
[Papers]
Gradient:
-
Searching for A Robust Neural Architecture in Four GPU Hours | [CVPR 2019]
- D-X-Y/GDAS | [Pytorch]
-
ASAP: Architecture Search, Anneal and Prune | [2019/04]
-
Single-Path NAS: Designing Hardware-Efficient ConvNets in less than 4 Hours | [2019/04]
- dstamoulis/single-path-nas | [Tensorflow]
-
Automatic Convolutional Neural Architecture Search for Image Classification Under Different Scenes | [IEEE Access 2019]
-
sharpDARTS: Faster and More Accurate Differentiable Architecture Search | [2019/03]
-
Learning Implicitly Recurrent CNNs Through Parameter Sharing | [ICLR 2019]
- lolemacs/soft-sharing | [Pytorch]
-
Probabilistic Neural Architecture Search | [2019/02]
-
Auto-DeepLab: Hierarchical Neural Architecture Search for Semantic Image Segmentation | [2019/01]
-
SNAS: Stochastic Neural Architecture Search | [ICLR 2019]
-
FBNet: Hardware-Aware Efficient ConvNet Design via Differentiable Neural Architecture Search | [2018/12]
-
Neural Architecture Optimization | [NIPS 2018]
- renqianluo/NAO | [Tensorflow]
- DARTS: Differentiable Architecture Search | [2018/06]
- quark0/darts | [Pytorch]
- khanrc/pt.darts | [Pytorch]
- dragen1860/DARTS-PyTorch | [Pytorch]
Reinforcement Learning:
-
Template-Based Automatic Search of Compact Semantic Segmentation Architectures | [2019/04]
-
Understanding Neural Architecture Search Techniques | [2019/03]
-
Fast, Accurate and Lightweight Super-Resolution with Neural Architecture Search | [2019/01]
- falsr/FALSR | [Tensorflow]
-
Multi-Objective Reinforced Evolution in Mobile Neural Architecture Search | [2019/01]
- moremnas/MoreMNAS | [Tensorflow]
-
ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware | [ICLR 2019]
- MIT-HAN-LAB/ProxylessNAS | [Pytorch, Tensorflow]
-
Transfer Learning with Neural AutoML | [NIPS 2018]
-
Learning Transferable Architectures for Scalable Image Recognition | [2018/07]
- wandering007/nasnet-pytorch | [Pytorch]
- tensorflow/models/research/slim/nets/nasnet | [Tensorflow]
-
MnasNet: Platform-Aware Neural Architecture Search for Mobile | [2018/07]
- AnjieZheng/MnasNet-PyTorch | [Pytorch]
-
Practical Block-wise Neural Network Architecture Generation | [CVPR 2018]
-
Efficient Neural Architecture Search via Parameter Sharing | [ICML 2018]
- melodyguan/enas | [Tensorflow]
- carpedm20/ENAS-pytorch | [Pytorch]
- Efficient Architecture Search by Network Transformation | [AAAI 2018]
Evolutionary Algorithm:
-
Single Path One-Shot Neural Architecture Search with Uniform Sampling | [2019/04]
-
DetNAS: Neural Architecture Search on Object Detection | [2019/03]
-
The Evolved Transformer | [2019/01]
-
Designing neural networks through neuroevolution | [Nature Machine Intelligence 2019]
-
EAT-NAS: Elastic Architecture Transfer for Accelerating Large-scale Neural Architecture Search | [2019/01]
- Efficient Multi-objective Neural Architecture Search via Lamarckian Evolution | [ICLR 2019]
SMBO:
-
MFAS: Multimodal Fusion Architecture Search | [CVPR 2019]
-
DPP-Net: Device-aware Progressive Search for Pareto-optimal Neural Architectures | [ECCV 2018]
- Progressive Neural Architecture Search | [ECCV 2018]
- titu1994/progressive-neural-architecture-search | [Keras, Tensorflow]
- chenxi116/PNASNet.pytorch | [Pytorch]
Random Search:
-
Exploring Randomly Wired Neural Networks for Image Recognition | [2019/04]
- Searching for Efficient Multi-Scale Architectures for Dense Image Prediction | [NIPS 2018]
Hypernetwork:
- Graph HyperNetworks for Neural Architecture Search | [ICLR 2019]
Bayesian Optimization:
Partial Order Pruning
- Partial Order Pruning: for Best Speed/Accuracy Trade-off in Neural Architecture Search | [CVPR 2019]
- lixincn2015/Partial-Order-Pruning | [Caffe]
Knowledge Distillation
[Projects]
- Microsoft/nni | [Python]
2.) 轻量级结构
[Papers]
Backbone:
- Searching for MobileNetV3 | [2019/05]
- kuan-wang/pytorch-mobilenet-v3 | [Pytorch]
- leaderj1001/MobileNetV3-Pytorch | [Pytorch]
Segmentation:
-
CGNet: A Light-weight Context Guided Network for Semantic Segmentation | [2019/04]
- wutianyiRosun/CGNet | [Pytorch]
-
ESPNetv2: A Light-weight, Power Efficient, and General Purpose Convolutional Neural Network | [2018/11]
- sacmehta/ESPNetv2 | [Pytorch]
-
ESPNet: Efficient Spatial Pyramid of Dilated Convolutions for Semantic Segmentation | [ECCV 2018]
- sacmehta/ESPNet | [Pytorch]
-
BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentation | [ECCV 2018]
- ooooverflow/BiSeNet | [Pytorch]
- ycszen/TorchSeg | [Pytorch]
- ERFNet: Efficient Residual Factorized ConvNet for Real-time Semantic Segmentation | [T-ITS 2017]
- Eromera/erfnet_pytorch | [Pytorch]
Object Detection:
-
ThunderNet: Towards Real-time Generic Object Detection | [2019/03]
-
Pooling Pyramid Network for Object Detection | [2018/09]
- tensorflow/models | [Tensorflow]
-
Tiny-DSOD: Lightweight Object Detection for Resource-Restricted Usages | [BMVC 2018]
- lyxok1/Tiny-DSOD | [Caffe]
-
Pelee: A Real-Time Object Detection System on Mobile Devices | [NeurIPS 2018]
- Robert-JunWang/Pelee | [Caffe]
- Robert-JunWang/PeleeNet | [Pytorch]
-
Receptive Field Block Net for Accurate and Fast Object Detection | [ECCV 2018]
- ruinmessi/RFBNet | [Pytorch]
- ShuangXieIrene/ssds.pytorch | [Pytorch]
- lzx1413/PytorchSSD | [Pytorch]
-
FSSD: Feature Fusion Single Shot Multibox Detector | [2017/12]
- ShuangXieIrene/ssds.pytorch | [Pytorch]
- lzx1413/PytorchSSD | [Pytorch]
- dlyldxwl/fssd.pytorch | [Pytorch]
- Feature Pyramid Networks for Object Detection | [CVPR 2017]
- tensorflow/models | [Tensorflow]
3.) 模型压缩和加速
[Papers]
Compression:
-
The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks | [ICLR 2019]
- google-research/lottery-ticket-hypothesis | [Tensorflow]
-
Rethinking the Value of Network Pruning | [ICLR 2019]
-
Slimmable Neural Networks | [ICLR 2019]
- JiahuiYu/slimmable_networks | [Pytorch]
-
AMC: AutoML for Model Compression and Acceleration on Mobile Devices | [ECCV 2018]
-
Learning Efficient Convolutional Networks through Network Slimming | [ICCV 2017]
- foolwood/pytorch-slimming | [Pytorch]
-
Channel Pruning for Accelerating Very Deep Neural Networks | [ICCV 2017]
- yihui-he/channel-pruning | [Caffe]
-
Pruning Convolutional Neural Networks for Resource Efficient Inference | [ICLR 2017]
- jacobgil/pytorch-pruning | [Pytorch]
- Pruning Filters for Efficient ConvNets | [ICLR 2017]
Acceleration:
- Fast Algorithms for Convolutional Neural Networks | [CVPR 2016]
- andravin/wincnn | [Python]
[Projects]
- NervanaSystems/distiller | [Pytorch]
- Tencent/PocketFlow | [Tensorflow]
[Tutorials/Blogs]
4.) 超参数优化
[Papers]
-
Tuning Hyperparameters without Grad Students: Scalable and Robust Bayesian Optimisation with Dragonfly | [2019/03]
-
Efficient High Dimensional Bayesian Optimization with Additivity and Quadrature Fourier Features | [NeurIPS 2018]
- Google vizier: A service for black-box optimization | [SIGKDD 2017]
[Projects]
- BoTorch | [PyTorch]
- Ax (Adaptive Experimentation Platform) | [PyTorch]
- Microsoft/nni | [Python]
- dragonfly/dragonfly | [Python]
[Tutorials/Blogs]
-
Hyperparameter tuning in Cloud Machine Learning Engine using Bayesian Optimization
- Bayesian optimization
- krasserm/bayesian-machine-learning | [Python]
5.) ，自动特征工程
Model Analyzer
-
Netscope CNN Analyzer | [Caffe]
-
sksq96/pytorch-summary | [Pytorch]
-
Lyken17/pytorch-OpCounter | [Pytorch]
- sovrasov/flops-counter.pytorch | [Pytorch]
References
- LITERATURE ON NEURAL ARCHITECTURE SEARCH
- handong1587/handong1587.github.io
- hibayesian/awesome-automl-papers
- mrgloom/awesome-semantic-segmentation
- amusi/awesome-object-detection
推荐阅读
- 基于 TensorFlow 的 AutoML 框架：谷歌开源 AdaNet
- Deep Compression/Acceleration（模型压缩加速论文汇总）
- 不用重新训练，直接将现有模型转换为 MobileNet
微信公众号: 极市平台（ID: extrememart ）
每天推送最新CV干货
自由转载-非商用-非衍生-保持署名（创意共享3.0许可证）
