Switch branches/tags
Nothing to show
Find file History
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
..
Failed to load latest commit information.
figures Update CIFAR10 results Sep 16, 2016
log Update CIFAR10 results Sep 16, 2016
README.md Adapt to keras 2.0.8 Oct 29, 2017
densenet.py fix BN axis error Dec 23, 2017
plot_results.py Improved plot visualisation. Nov 10, 2017
run_cifar10.py Adapt to keras 2.0.8 Oct 29, 2017

README.md

Keras Implementation of DenseNet

Original idea and implementation:

Densely Connected Convolutional Network

The figures below are taken from the paper above.

Dense block

Figure 1: A dense block with 5 layers and growth rate 4.

Model scheme

Figure 2: A deep DenseNet with three dense blocks.

Results:

Below, results obtained with a network of depth 40, growth rate 12, 3 dense blocks, dropout rate of 0.2 and trained with SGD for 276 epochs.

All convolutional layer have bias = False meaning we don't use a bias parameter for them.

Weight decay (1E-4) is applied to convolutional layers, batch norm parameters and the last dense layer.

The initial learning rate is 0.1 and the learning rate is divided by 10 after 150 and 225 epochs.

These settings lead to the same results as Densely Connected Convolutional Network: 7 % misclassification rate on the CIFAR10 test set without data augmentation.

Model scheme

Running a CIFAR10 experiment

python run_cifar10.py

optional arguments:

Usage guide:

python run_cifar10.py

optional arguments:

-h, --help show this help message and exit
--batch_size BATCH_SIZE Batch size
--nb_epoch NB_EPOCH  Number of epochs
--depth DEPTH  Network depth
--nb_dense_block NB_DENSE_BLOCK Number of dense blocks
--nb_filter NB_FILTER Initial number of conv filters
--growth_rate GROWTH_RATE Number of new filters added by conv layers
--dropout_rate DROPOUT_RATE  Dropout rate
--learning_rate LEARNING_RATE Learning rate
--weight_decay WEIGHT_DECAY L2 regularization on weights
--plot_architecture PLOT_ARCHITECTURE Save a plot of the network architecture

Architecture

With two dense blocks and 2 convolution operations within each block, the model looks like this:

Model archi

Requirements

  • numpy==1.13.3
  • matplotlib==2.0.2
  • Keras==2.0.8
  • tensorflow==1.3.0 or theano==0.9.0