Welcome to the SimbaML Documentation!#

Date: Oct 10, 2023 Version: 1.0.0rc14+1.gdba84d0

SimbaML is an all-in-one framework for integrating prior knowledge of ODE models into the ML process by synthetic data augmentation. It allows for the convenient generation of realistic synthetic data by sparsifying and adding noise. Furthermore, our framework provides customizable pipelines for various ML experiments, such as identifying needs for data collection and transfer learning.

Visual Abstract of SimbaML

Useful links: Installation | Source Repository | Issue Tracker | Mailing List

Installation#

SimbaML requires Python 3.10 or newer and can be installed via pip:

pip install simba_ml

You can check if the installation was successful by importing the package and checking the version:

import simba_ml
simba_ml.__version__

For more detailed installation instructions and requirements, see Installation.

For an example usage of SimbaML, see Quickstart.

Reference#

When using SimbaML in a scientific publication, please include the following references to relevant papers.

@inproceedings{DBLP:conf/iclr/KleisslDHZIWRB23,
    author       = {Maximilian Kleissl and
                    Lukas Drews and
                    Benedict B. Heyder and
                    Julian Zabbarov and
                    Pascal Iversen and
                    Simon Witzke and
                    Bernhard Y. Renard and
                    Katharina Baum},
    editor       = {Krystal Maughan and
                    Rosanne Liu and
                    Thomas F. Burns},
    title        = {SimbaML: Connecting Mechanistic Models and Machine Learning with Augmented
                    Data},
    booktitle    = {The First Tiny Papers Track at {ICLR} 2023, Tiny Papers @ {ICLR} 2023,
                    Kigali, Rwanda, May 5, 2023},
    publisher    = {OpenReview.net},
    year         = {2023},
    url          = {https://openreview.net/pdf?id=1wtUadpmVzu},
    timestamp    = {Wed, 19 Jul 2023 17:21:16 +0200},
    biburl       = {https://dblp.org/rec/conf/iclr/KleisslDHZIWRB23.bib},
    bibsource    = {dblp computer science bibliography, https://dblp.org}
}