SchNetPack: A Deep Learning Toolbox For Atomistic Systems.
By Kristof T. Schütt, Pan Kessel, Michael Gastegger, K. A. Nicoli, Alexandre Tkatchenko, K. Müller. 2018
389 citations recorded by the source corpus.
SchNetPack is a toolbox for the development and application of deep neural networks that predict potential energy surfaces and other quantum-chemical properties of molecules and materials. It contains basic building blocks of atomistic neural networks, manages their training, and provides simple access to common benchmark datasets. This allows for an easy implementation and evaluation of new models. For now, SchNetPack includes implementations of (weighted) atom-centered symmetry functions and the deep tensor neural network SchNet, as well as ready-to-use scripts that allow one to train these models on molecule and material datasets. Based on the PyTorch deep learning framework, SchNetPack allows one to efficiently apply the neural networks to large datasets with millions of reference calculations, as well as parallelize the model across multiple GPUs. Finally, SchNetPack provides an interface to the Atomic Simulation Environment in order to make trained models easily accessible to researchers that are not yet familiar with neural networks.
Subfields: Computational Theory and Mathematics, Materials Chemistry
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computational Theory and Mathematics
- Tess E. Smidt
- Frank No'e
- Mario Geiger
- Michael Gastegger
- Grgoire Montavon
- Kristof T. Schütt
- Klaus‐Robert Müller
- Florian Hase
In Materials Chemistry
- Frank No'e
- Tess E. Smidt
- Michael Gastegger
- Mario Geiger
- Klaus‐Robert Müller
- Cecilia Clementi
- Kristof T. Schütt
- Grgoire Montavon
Citations in the corpus
- Relevance of Rotationally Equivariant Convolutions for Predicting Molecular Properties (2020, methodology): “This is easily done by modifying the tutorial example in SchNetPack [13].”
- Recent advances and applications of deep learning methods in materials science (2021, neutral): “com/atomistic-machine-learning [94] CGCNN Sol https://github.”
- Deep Potentials for Materials Science (2022, methodology): “[24, 25] developed SchNet and the SchNetPack package based upon a neural network framework to model the chemical properties and PES of molecular materials.”
- Machine-learned potentials for next-generation matter simulations (2021, methodology): “The ANI1 18 model as well as SchNetPack 143 and sGDML (http:// quantum-machine.org/gdml) provide interfaces to the atomic simulation environment (ASE).”
- Inverse design of 3d molecular structures with conditional generative neural networks (2021, methodology): “We use pretrained SchNet21,67 models from SchNetPack68 to predict the HOMO-LUMO gap, isotropic polarizability, and internal energy at zero Kelvin of generated molecules.”