ID	densenet121
LR	0.1
Epochs	90
LR Gamma	0.1
Momentum	0.9
Batch Size	32
LR Step Size	30
Weight Decay	0.0001

ID	densenet161
LR	0.1
Epochs	90
LR Gamma	0.1
Momentum	0.9
Batch Size	32
LR Step Size	30
Weight Decay	0.0001

ID	densenet169
LR	0.1
Epochs	90
LR Gamma	0.1
Momentum	0.9
Batch Size	32
LR Step Size	30
Weight Decay	0.0001

ID	densenet201
LR	0.1
Epochs	90
LR Gamma	0.1
Momentum	0.9
Batch Size	32
LR Step Size	30
Weight Decay	0.0001

DenseNet

pytorch / vision

Last updated on Feb 12, 2021

Parameters 8 Million

FLOPs 3 Billion

File Size 30.84 MB

Training Data ImageNet

Training Resources 8x NVIDIA V100 GPUs

Training Time

Training Techniques	Weight Decay, SGD with Momentum
Architecture	1x1 Convolution, Average Pooling, Batch Normalization, Convolution, Dense Block, Dropout, Dense Connections, ReLU, Max Pooling, Softmax
ID	densenet121
LR	0.1
Epochs	90
LR Gamma	0.1
Momentum	0.9
Batch Size	32
LR Step Size	30
Weight Decay	0.0001
SHOW MORE
SHOW LESS

Parameters 29 Million

FLOPs 8 Billion

File Size 110.37 MB

Training Data ImageNet

Training Resources 8x NVIDIA V100 GPUs

Training Time

Training Techniques	Weight Decay, SGD with Momentum
Architecture	1x1 Convolution, Average Pooling, Batch Normalization, Convolution, Dense Block, Dropout, Dense Connections, ReLU, Max Pooling, Softmax
ID	densenet161
LR	0.1
Epochs	90
LR Gamma	0.1
Momentum	0.9
Batch Size	32
LR Step Size	30
Weight Decay	0.0001
SHOW MORE
SHOW LESS

Parameters 14 Million

FLOPs 3 Billion

File Size 54.71 MB

Training Data ImageNet

Training Resources 8x NVIDIA V100 GPUs

Training Time

Training Techniques	Weight Decay, SGD with Momentum
Architecture	1x1 Convolution, Average Pooling, Batch Normalization, Convolution, Dense Block, Dropout, Dense Connections, ReLU, Max Pooling, Softmax
ID	densenet169
LR	0.1
Epochs	90
LR Gamma	0.1
Momentum	0.9
Batch Size	32
LR Step Size	30
Weight Decay	0.0001
SHOW MORE
SHOW LESS

Parameters 20 Million

FLOPs 4 Billion

File Size 77.37 MB

Training Data ImageNet

Training Resources 8x NVIDIA V100 GPUs

Training Time

Training Techniques	Weight Decay, SGD with Momentum
Architecture	1x1 Convolution, Average Pooling, Batch Normalization, Convolution, Dense Block, Dropout, Dense Connections, ReLU, Max Pooling, Softmax
ID	densenet201
LR	0.1
Epochs	90
LR Gamma	0.1
Momentum	0.9
Batch Size	32
LR Step Size	30
Weight Decay	0.0001
SHOW MORE
SHOW LESS

README.md

Summary

DenseNet is a type of convolutional neural network that utilises dense connections between layers, through Dense Blocks, where we connect all layers (with matching feature-map sizes) directly with each other. To preserve the feed-forward nature, each layer obtains additional inputs from all preceding layers and passes on its own feature-maps to all subsequent layers.

How do I load this model?

To load a pretrained model:

import torchvision.models as models
densenet = models.densenet161(pretrained=True)

Replace the model name with the variant you want to use, e.g. densenet161 You can find the IDs in the model summaries at the top of this page.

To evaluate the model, use the image classification recipes from the library.

python train.py --test-only --model='<model_name>'

How do I train this model?

You can follow the torchvision recipe on GitHub for training a new model afresh.

Citation

@article{DBLP:journals/corr/HuangLW16a,
  author    = {Gao Huang and
               Zhuang Liu and
               Kilian Q. Weinberger},
  title     = {Densely Connected Convolutional Networks},
  journal   = {CoRR},
  volume    = {abs/1608.06993},
  year      = {2016},
  url       = {http://arxiv.org/abs/1608.06993},
  archivePrefix = {arXiv},
  eprint    = {1608.06993},
  timestamp = {Mon, 10 Sep 2018 15:49:32 +0200},
  biburl    = {https://dblp.org/rec/journals/corr/HuangLW16a.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}

Results

Image Classification on ImageNet

MODEL	TOP 1 ACCURACY	TOP 5 ACCURACY
Densenet-161	77.65%	93.8%
Densenet-201	77.2%	93.57%
Densenet-169	76.0%	93.0%
Densenet-121	74.65%	92.17%