soheilabadifard / DynED

Dynamic Ensemble Diversification

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

DeepSource

DynED

The datasets that DynED is evaluated on are available in the Dataset folder. To run the DynED, please install the Scikit-multiflow package. So that you know, the last version of scikit-multiflow is compatible with Python 3.8 and requires a specific version of Numpy and Pandas Packages. All the installation instructions are provided on the scikit-multiflow home website: https://scikit-multiflow.readthedocs.io/en/stable/index.html.

Then, open the DynED.py file with any IDE or Jupyter Notebook and simply replace the full path of the dataset you are trying to evaluate with "Put the full address and name of the dataset here." Next, run the Python Script. It will output the Mean accuracy of DynED on the provided dataset.

BaseLine Experiments

Please take a look at the README.md in the scripts directory.

Acknowledgments

Tübitak supports this project under grant 122E271.

About

Dynamic Ensemble Diversification

License:GNU Affero General Public License v3.0


Languages

Language:Python 100.0%