QML: A Python Toolkit for Quantum Machine Learning

QML is a Python2/3-compatible toolkit for representation learning of properties of molecules and solids. QML is not a high-level framework where you can do model.train(), but supplies the building blocks to carry out efficient and accurate machine learning on chemical compounds. As such, the goal is to provide usable and efficient implementations of concepts such as representations and kernels.

Current list of contributors:

• Anders S. Christensen (University of Basel)
• Lars A. Bratholm (University of Bristol)
• Jimmy C. Kromann (University of Basel)
• Silvia Amabilino (University of Bristol)
• Felix A. Faber (University of Basel)
• Bing Huang (University of Basel)
• Alexandre Tkatchenko (University of Luxembourg)
• Klaus-Robert Muller (Technische Universität Berlin/Korea University)
• David R. Glowacki (University of Bristol)
• 15. Anatole von Lilienfeld (University of Basel)

Code development

The QML code is developed through our GitHub repository:

https://github.com/qmlcode/qml

Please add you code to QML by forking and making pull-requests to the “develop” branch. Every now and then develop branch is pushed to the “master” branch and automatically deployed to PyPI, where the latest stable version is hosted.

See the “Installing QML” page for up-to-date installation instructions.

Citing QML:

Until the preprint is available from arXiv, please cite use of QML as:

AS Christensen, FA Faber, B Huang, LA Bratholm, A Tkatchenko, KR Muller, OA von Lilienfeld (2017) "QML: A Python Toolkit for Quantum Machine Learning" https://github.com/qmlcode/qml

For citation of the individual procedures of QML, please see the “Citing use of QML” section.

GETTING STARTED:

• Installing QML
  • Installing via pip
  • Installing via with Intel compilers
  • Note on Apple/Mac support
  • Report Bugs
• Citing use of QML
• QML Tutorial
  • Theory
  • Tutorial exercises
  • Exercise 1: Representations
  • Exercise 2: Kernels
  • Exercise 3: Regression
  • Exercise 4: Prediction
  • Exercise 5: Learning curves
  • Exercise 6: Delta learning
  • References
• Examples
  • Generating representations using the Compound class
  • Generating representations via the qml.representations module
  • Calculating a Gaussian kernel
  • Calculating a Gaussian kernel using a local representation
  • Generating the SLATM representation
  • Generating the FCHL representation
  • Generating the FCHL kernel

SOURCE DOCUMENTATION:

• Python API documentation
  • qml.representations module
  • qml.kernels module
  • qml.distance module
  • qml.math module
  • qml.Compound class
  • qml.fchl module
  • qml.wrappers module

License:

QML is freely available under the terms of the MIT license.