IOKR

This open source Python library provide several methods for Output Kernelization.

Website and documentation: https://IOKR.github.io/

Source Code (MIT): https://github.com/IOKR/IOKR

Installation

The library has been tested on Linux, MacOSX and Windows. It requires a C++ compiler for building/installing the EMD solver and relies on the following Python modules:

Pandas (>=1.2)
Numpy (>=1.16)
Scipy (>=1.0)
Scikit-learn (>=1.0)

Pip installation

You can install the toolbox through PyPI with:

pip install IOKR

Anaconda installation with conda-forge

If you use the Anaconda python distribution, POT is available in conda-forge. To install it and the required dependencies:

conda install -c conda-forge IOKR

Post installation check

After a correct installation, you should be able to import the module without errors:

import IOKR

Examples

Short examples

Import the toolbox

import IOKR

Run IOKR

from IOKR.model.model import IOKR
from sklearn.model_selection import train_test_split
from IOKR.data.load_data import load_bibtex
from sklearn.metrics import f1_score

path = "IOKR/data/bibtex"
X, Y, _, _ = load_bibtex(path)

X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.33, random_state=42)

clf = IOKR()
clf.verbose = 1
L = 1e-5
sx = 1000
sy = 10

clf.fit(X=X_train, Y=Y_train, L=L, sx=sx, sy=sy)
Y_pred_train = clf.predict(X_test=X_train)
Y_pred_test = clf.predict(X_test=X_test)
f1_train = f1_score(Y_pred_train, Y_train, average='samples')
f1_test = f1_score(Y_pred_test, Y_test, average='samples')
print("Train f1 score:", f1_train,"/", "Test f1 score:", f1_test)

Examples and Notebooks

The examples folder contain several examples and use case for the library. The full documentation with examples and output is available on https://IOKR.github.io/.

Acknowledgements

This toolbox has been created and is maintained by

Hi! PARIS

The contributors to this library are

Florence d'Alché-Buc (Researcher)
Luc Motte (Researcher)
Tamim El Ahmad (Researcher)
Gaëtan Brison (Engineer)
Danaël Schlewer-Becker (Engineer)
Awais Sani (Engineer)

Contributions and code of conduct

Every contribution is welcome and should respect the contribution guidelines. Each member of the project is expected to follow the code of conduct.

Support

You can ask questions and join the development discussion:

On the IOKR slack channel
On the IOKR gitter channel
On the IOKR mailing list

You can also post bug reports and feature requests in Github issues. Make sure to read our guidelines first.

References

[1] Brouard, C., d'Alché-Buc, F., Szafranski, M. (2013, November). Semi-supervised Penalized Output Kernel Regression for Link Prediction. 28th International Conference on Machine Learning (ICML 2011), pp.593–600.

[2] Brouard, C., Szafranski, M., d'Alché-Buc, F. (2016). Input Output Kernel Regression: Supervised and Semi-Supervised Structured Output Prediction with Operator-Valued Kernels. Journal of Machine Learning Research, Microtome Publishing.

[3] Brouard, C., Shen, H., Dührkop, K., d'Alché-Buc, F., Böcker, S., & Rousu, J. (2016) Fast metabolite identification with input output kernel regression. Bioinformatics, Oxford University Press (OUP)

hi-paris / IOKR