Commit 7f1f66ac authored by Florent Chatelain's avatar Florent Chatelain
Browse files

arxiv suppmat

parent c67fc5ad
......@@ -4,16 +4,16 @@ This is the companion gitlab repository for our [arXiv manuscript](https://arxiv
## Supplementary materials
### Proofs
### Proofs
The elements of proof of the main theorems in the core article are detailled in the pdf file
The elements of proof of the main theorems in the core article are detailed in the pdf file
[supp_mat_proofs.pdf](https://gricad-gitlab.univ-grenoble-alpes.fr/chatelaf/two-way-kernel-matrix-puncturing/-/blob/master/supp_mat_proofs.pdf)
### Code
The python jupyter notebook [supp_mat_figures.ipynb](https://gricad-gitlab.univ-grenoble-alpes.fr/chatelaf/two-way-kernel-matrix-puncturing/-/blob/master/supp_mat_figures.ipynb) is set up to replicate the figures shown in the paper.
This loads the `punctutils.py` function module and the GAN data. The python modules needed to run the notebook are listed in the first cells.
This loads the `punctutils.py` function module and the GAN data. The python modules needed to run the notebook include notably `seaborn` for plotting or `tensorflow` for loading the MNIST-fashion dataset (Figs 6-8).
#### Requirements
......
jupyter labextension install --minimize=False @jupyter-widgets/jupyterlab-manager
numpy==1.18.1
scipy==1.4.1
matplotlib==3.3.3
scikit-learn==0.23.2
pandas==1.0.3
seaborn==0.11.1
tensorflow==2.1.0
%% Cell type:markdown id: tags:
#
This notebook can be run with the binder: [![Binder](https://mybinder.org/badge_logo.svg)](https://mybinder.org/v2/git/https%3A%2F%2Fgricad-gitlab.univ-grenoble-alpes.fr%2Fchatelaf%2Ftwo-way-kernel-matrix-puncturing/master?urlpath=lab/tree/supp_mat_figures.ipynb)
# <center>*"Two-way kernel matrix puncturing: towards resource-efficient PCA and spectral clustering"*</center>
## <center>-- Supplementary Material -- </center>
## <center>-- Python Codes of main article figures --</center>
## Preamble: useful packages and functions
......@@ -372,11 +373,11 @@
%% Cell type:markdown id: tags:
#### Figure 6
Examples of **MNIST-fashion images**, `trouser` instances **(top row)**, `pullover` instances **(bottom row)**.
**Note:** the GAN data we generated and used in the submitted paper are too voluminous to be included in the supplementary material and is not publicly and anonymously available. For this reason, we are using below the smaller, publicly available, MNIST-fashion real word data set. However **the conclusions obtained for this dataset are very similar to those drawn in the paper for the GAN data.**
**Note:** the GAN data we generated and used in the submitted paper are quite vomuminous. As an alternative we are using below the smaller, publicly available, MNIST-fashion real word data set. However **the conclusions obtained for this dataset are very similar to those drawn in the paper for the GAN data.**
%% Cell type:code id: tags:
``` python
from tensorflow.keras.datasets import fashion_mnist
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment