Re-implementation of Convolutional Kernel Network (CKN) from Mairal (2016) in Python based on the Pytorch framework. The package is available under the GPL-v3 license.
Author: Dexiong Chen
Credits: Ghislain Durif, Mathilde Caron, Alberto Bietti, Julien Mairal
The code is based on
Mairal, Julien. End-to-end kernel learning with supervised convolutional kernel networks. NIPS 2016.
If you have any issues, please contact dexiong.chen@inria.fr.
We strongly recommend users to use anaconda to install the following packages
numpy
scipy
scikit-learn
pytorch=1.2.0
miso_svm
The Python package miso_svm can be installed with (original repository)
cd third-party/miso_svm-1.0
python setup.py install
Reproduction of the results from Mairal (2016) with this package.
The results from the original paper (Mairal, 2016) were achieved using
cudnn-based Matlab code available here. To run the following experiments, please first download the data, put into the folder ./data/cifar-10 and then do
export PYTHONPATH=$PWD:$PYTHONPATH
cd experimentsHere is a summary of the results of unsupervised CKN on CIFAR10 image classification dataset with pre-whitening and without data augmentation or model ensembling.
# Code examples
python cifar10_unsup.py --filters 64 256 --subsamplings 2 6 --kernel-sizes 3 3| #layers | #filters | filter size | subsampling | sigma | Accuracy |
|---|---|---|---|---|---|
| 2 | 64, 256 | 3, 3 | 2, 6 | 0.6 | 77.5 |
| 2 | 256, 1024 | 3, 3 | 2, 6 | 0.6 | 82.0 |
| 2 | 512, 8192 | 3, 2 | 2, 6 | 0.6 | 84.0 |
Here is a summary of the results of supervised CKN on CIFAR10 image classification dataset with pre-whitening and without data augmentation or model ensembling.
# Code examples
python cifar10_sup.py --epochs 105 --lr 0.1 --alpha 0.001 --loss hinge --alternating --model ckn5
python cifar10_sup.py --epochs 105 --lr 0.1 --alpha 0.1 --loss hinge --alternating --model ckn14| Architecture | Accuracy | training time (GTX1080_ti) |
|---|---|---|
| CKN-5 | 86.1 | ~60 min |
| CKN-14 | 90.2 | ~260 min |