模型概述
包含的模型架构来自各种来源。来源包括论文、我重写/改编的原始实现(“参考代码”)以及我直接利用的 PyTorch 实现(“代码”),如下所示。
大多数包含的模型都具有预训练权重。权重要么是
- 来自其原始来源的
- 由我自己从其在不同框架中的原始实现移植而来(例如 Tensorflow 模型)
- 使用包含的训练脚本从头开始训练的
预训练权重的验证结果在此
关于 timm
中模型的更精彩视图(附带精美图片)可以在paperswithcode中找到。
Big Transfer ResNetV2 (BiT)
- 实现:resnetv2.py
- 论文:
Big Transfer (BiT): General Visual Representation Learning
- https://arxiv.org/abs/1912.11370 - 参考代码:https://github.com/google-research/big_transfer
跨阶段局部网络
- 实现:cspnet.py
- 论文:
CSPNet: A New Backbone that can Enhance Learning Capability of CNN
- https://arxiv.org/abs/1911.11929 - 参考实现:https://github.com/WongKinYiu/CrossStagePartialNetworks
DenseNet
- 实现:densenet.py
- 论文:
Densely Connected Convolutional Networks
- https://arxiv.org/abs/1608.06993 - 代码:https://github.com/pytorch/vision/tree/master/torchvision/models
DLA
双路径网络
- 实现:dpn.py
- 论文:
Dual Path Networks
- https://arxiv.org/abs/1707.01629 - 我的 PyTorch 代码:https://github.com/rwightman/pytorch-dpn-pretrained
- 参考代码:https://github.com/cypw/DPNs
GPU 高效网络
- 实现:byobnet.py
- 论文:
Neural Architecture Design for GPU-Efficient Networks
- https://arxiv.org/abs/2006.14090 - 参考代码:https://github.com/idstcv/GPU-Efficient-Networks
HRNet
- 实现:hrnet.py
- 论文:
Deep High-Resolution Representation Learning for Visual Recognition
- https://arxiv.org/abs/1908.07919 - 代码:https://github.com/HRNet/HRNet-Image-Classification
Inception-V3
- 实现:inception_v3.py
- 论文:
Rethinking the Inception Architecture for Computer Vision
- https://arxiv.org/abs/1512.00567 - 代码:https://github.com/pytorch/vision/tree/master/torchvision/models
Inception-V4
- 实现:inception_v4.py
- 论文:
Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning
- https://arxiv.org/abs/1602.07261 - 代码:https://github.com/Cadene/pretrained-models.pytorch
- 参考代码:https://github.com/tensorflow/models/tree/master/research/slim/nets
Inception-ResNet-V2
- 实现:inception_resnet_v2.py
- 论文:
Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning
- https://arxiv.org/abs/1602.07261 - 代码:https://github.com/Cadene/pretrained-models.pytorch
- 参考代码:https://github.com/tensorflow/models/tree/master/research/slim/nets
NASNet-A
- 实现:nasnet.py
- 论文:
Learning Transferable Architectures for Scalable Image Recognition
- https://arxiv.org/abs/1707.07012 - 代码:https://github.com/Cadene/pretrained-models.pytorch
- 参考代码:https://github.com/tensorflow/models/tree/master/research/slim/nets/nasnet
PNasNet-5
- 实施: pnasnet.py
- 论文:
Progressive Neural Architecture Search
- https://arxiv.org/abs/1712.00559 - 代码:https://github.com/Cadene/pretrained-models.pytorch
- 参考代码:https://github.com/tensorflow/models/tree/master/research/slim/nets/nasnet
EfficientNet
- 实施: efficientnet.py
- 论文
- EfficientNet NoisyStudent (B0-B7, L2) - https://arxiv.org/abs/1911.04252
- EfficientNet AdvProp (B0-B8) - https://arxiv.org/abs/1911.09665
- EfficientNet (B0-B7) - https://arxiv.org/abs/1905.11946
- EfficientNet-EdgeTPU (S, M, L) - https://ai.googleblog.com/2019/08/efficientnet-edgetpu-creating.html
- MixNet - https://arxiv.org/abs/1907.09595
- MNASNet B1, A1 (Squeeze-Excite), and Small - https://arxiv.org/abs/1807.11626
- MobileNet-V2 - https://arxiv.org/abs/1801.04381
- FBNet-C - https://arxiv.org/abs/1812.03443
- Single-Path NAS - https://arxiv.org/abs/1904.02877
- 我的 PyTorch 代码: https://github.com/rwightman/gen-efficientnet-pytorch
- 参考代码: https://github.com/tensorflow/tpu/tree/master/models/official/efficientnet
MobileNet-V3
- 实施: mobilenetv3.py
- 论文:
Searching for MobileNetV3
- https://arxiv.org/abs/1905.02244 - 参考代码: https://github.com/tensorflow/models/tree/master/research/slim/nets/mobilenet
RegNet
- 实施: regnet.py
- 论文:
Designing Network Design Spaces
- https://arxiv.org/abs/2003.13678 - 参考代码: https://github.com/facebookresearch/pycls/blob/master/pycls/models/regnet.py
RepVGG
- 实现:byobnet.py
- 论文:
Making VGG-style ConvNets Great Again
- https://arxiv.org/abs/2101.03697 - 参考代码: https://github.com/DingXiaoH/RepVGG
ResNet, ResNeXt
实施: resnet.py
ResNet (V1B)
- 论文:
Deep Residual Learning for Image Recognition
- https://arxiv.org/abs/1512.03385 - 代码:https://github.com/pytorch/vision/tree/master/torchvision/models
- 论文:
ResNeXt
- 论文:
Aggregated Residual Transformations for Deep Neural Networks
- https://arxiv.org/abs/1611.05431 - 代码:https://github.com/pytorch/vision/tree/master/torchvision/models
- 论文:
‘Bag of Tricks’ / Gluon C, D, E, S ResNet 变体
- 论文:
Bag of Tricks for Image Classification with CNNs
- https://arxiv.org/abs/1812.01187 - 代码: https://github.com/dmlc/gluon-cv/blob/master/gluoncv/model_zoo/resnetv1b.py
- 论文:
Instagram 预训练 / ImageNet 调优 ResNeXt101
- 论文:
Exploring the Limits of Weakly Supervised Pretraining
- https://arxiv.org/abs/1805.00932 - 权重: https://pytorch.ac.cn/hub/facebookresearch_WSL-Images_resnext (注意: CC BY-NC 4.0 许可证,对商业用途不友好)
- 论文:
半监督 (SSL) / 半弱监督 (SWSL) ResNet 和 ResNeXt
- 论文:
Billion-scale semi-supervised learning for image classification
- https://arxiv.org/abs/1905.00546 - 权重: https://github.com/facebookresearch/semi-supervised-ImageNet1K-models (注意: CC BY-NC 4.0 许可证,对商业用途不友好)
- 论文:
Squeeze-and-Excitation Networks
- 论文:
Squeeze-and-Excitation Networks
- https://arxiv.org/abs/1709.01507 - 代码: 已添加到 ResNet 基础架构中,这是当前的最新版本,旧的
senet.py
正在被弃用
- 论文:
ECAResNet (ECA-Net)
- 论文:
ECA-Net: Efficient Channel Attention for Deep CNN
- https://arxiv.org/abs/1910.03151v4 - 代码: 已添加到 ResNet 基础架构中,ECA 模块由 @VRandme 贡献,参考 https://github.com/BangguWu/ECANet
- 论文:
Res2Net
- 实施: res2net.py
- 论文:
Res2Net: A New Multi-scale Backbone Architecture
- https://arxiv.org/abs/1904.01169 - 代码: https://github.com/gasvn/Res2Net
ResNeSt
- 实施: resnest.py
- 论文:
ResNeSt: Split-Attention Networks
- https://arxiv.org/abs/2004.08955 - 代码: https://github.com/zhanghang1989/ResNeSt
ReXNet
- 实施: rexnet.py
- 论文:
ReXNet: Diminishing Representational Bottleneck on CNN
- https://arxiv.org/abs/2007.00992 - 代码: https://github.com/clovaai/rexnet
Selective-Kernel Networks
- 实施: sknet.py
- 论文:
Selective-Kernel Networks
- https://arxiv.org/abs/1903.06586 - 代码: https://github.com/implus/SKNet, https://github.com/clovaai/assembled-cnn
SelecSLS
- 实施: selecsls.py
- 论文:
XNect: Real-time Multi-Person 3D Motion Capture with a Single RGB Camera
- https://arxiv.org/abs/1907.00837 - 代码: https://github.com/mehtadushy/SelecSLS-Pytorch
Squeeze-and-Excitation Networks
实施: senet.py 注意: 我正在弃用此版本的网络,新版本是
resnet.py
的一部分论文:
Squeeze-and-Excitation Networks
- https://arxiv.org/abs/1709.01507
TResNet
- 实施: tresnet.py
- 论文:
TResNet: High Performance GPU-Dedicated Architecture
- https://arxiv.org/abs/2003.13630 - 代码: https://github.com/mrT23/TResNet
VGG
- 实施: vgg.py
- 论文:
Very Deep Convolutional Networks For Large-Scale Image Recognition
- https://arxiv.org/pdf/1409.1556.pdf - 参考代码: https://github.com/pytorch/vision/blob/master/torchvision/models/vgg.py
Vision Transformer
- 实施: vision_transformer.py
- 论文:
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
- https://arxiv.org/abs/2010.11929 - 参考代码和预训练权重: https://github.com/google-research/vision_transformer
VovNet V2 和 V1
- 实施: vovnet.py
- 论文:
CenterMask : Real-Time Anchor-Free Instance Segmentation
- https://arxiv.org/abs/1911.06667 - 参考代码: https://github.com/youngwanLEE/vovnet-detectron2
Xception
- 实施: xception.py
- 论文:
Xception: Deep Learning with Depthwise Separable Convolutions
- https://arxiv.org/abs/1610.02357 - 代码:https://github.com/Cadene/pretrained-models.pytorch
Xception (Modified Aligned, Gluon)
- 实施: gluon_xception.py
- 论文:
Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation
- https://arxiv.org/abs/1802.02611 - 参考代码: https://github.com/dmlc/gluon-cv/tree/master/gluoncv/model_zoo, https://github.com/jfzhang95/pytorch-deeplab-xception/
Xception (Modified Aligned, TF)
- 实施: aligned_xception.py
- 论文:
Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation
- https://arxiv.org/abs/1802.02611 - 参考代码: https://github.com/tensorflow/models/tree/master/research/deeplab