Awesome

PL-CNN

An optimization algorithm for learning Piecewise Linear Convolutional Neural Networks. This is the implementation of the paper Trusting SVM for Piecewise Linear CNNs by Leonard Berrada, Andrew Zisserman and M. Pawan Kumar.

Requirements

This code has been developped python 2.7 and requires the packages given in requirements.txt.

Repository organization

data

Contains the data set folders. The use of the following data sets is implemented:

• MNIST
• CIFAR-10
• CIFAR-100
• ImageNet

src

This contains the python code. To run the tests, run sh scripts/test.sh from the src working directory.

log

This contains the logs from the experiments as well as the saved models. All experiment logs are backed up in log/all.

To reproduce results

First make sure your working directory is src, and that the standard data sets are in data. Then to run experiments with gpu [x] with results output to directory log/my_dir:

MNIST:

sh scripts/mnist.sh gpu[x] my_dir

CIFAR-10:

sh scripts/cifar10.sh gpu[x] my_dir

CIFAR-100:

sh scripts/cifar100.sh gpu[x] my_dir

ImageNet:

The following code assumes that:

• the images have been extracted and resized to a minimal size of 256, and are in the path specified in datasets/imagenet.py.
• the pre-trained VGG-16 model has been downloaded from Lasagne model zoo and placed in my_dir, the directory for the experiment.

Launch data pipeline for training set: python server.py --train

And in a different shell: python lwsvm.py --dataset imagenet --in_name vgg16 --device gpu[x] --xp_dir my_dir

When training is complete, launch data pipeline for validation set: python server.py --val

And launch evaluation in a different shell: python extract.py --imagenet --weights_file imagenet_lwsvm_15

To re-use code

The architectures can easily be changed through the addLayer-type methods of NeuralNet. Note that the code assumes a restricted search on the conditional gradient of convolutional layers (Observation 1 in section 4 of the paper), since they are usually followed by an important number of non-linearities (max-pooling).