De novo, in-silico design of molecules is a challenging problem with applications in drug discovery and material design. We introduce a masked graph model, which learns a distribution over graphs by capturing conditional distributions over unobserved nodes (atoms) and edges (bonds) given observed ones. We train and then sample from …
read moreCASM interfaces with first-principles electronic structure codes, automates the construction and parameterization of effective Hamiltonians and subsequently builds highly optimized (kinetic) Monte Carlo codes to predict finite-temperature thermodynamic and kinetic properties. CASM uses group theoretic techniques that take full advantage of crystal symmetry in order to rigorously construct effective Hamiltonians …
read moreCrySPY enables anyone to easily perform crystal structure prediction simulations for materials discovery and design, and automates structure generation, structure optimization, energy evaluation, and efficiently selecting candidates using machine learning. Several searching algorithms are available such as random search, evolutionary algorithm, Bayesian optimization, and Look Ahead based on Quadratic Approximation. …
read moreThe atomic simulation environment (ASE) is a software package written in the Python programming language with the aim of setting up, steering, and analyzing atomistic simulations. In ASE, tasks are fully scripted in Python. The powerful syntax of Python combined with the NumPy array library make it possible to perform …
read moreThe code is freely available under the GNU LGPL license. The Atoms object Working with the array methods of Atoms objects Unit cell and boundary conditions Special attributes Adding a calculator List-methods Other methods List of all Methods The Cell object Units Changing the CODATA version File input and output …
read moreA particular challenge for new materials is that the atomic positions are generally not known. We present a machine learning model for the prediction of DFT-calculated formation energies based on Voronoi quotient graphs and local symmetry classification without the need for detailed information about atomic positions.
read moreTensorflow implementation of message passing neural networks for molecules and materials. The framework implements the SchNet model and its extension with edge update network NMP-EDGE as well as the model used in Materials property prediction using symmetry-labeled graphs as atomic-position independent descriptors. The main difference between msgnet and schnetpack is …
read moreGilmer, et al.(Neural Message Passing) investigated various graph neural networks for predicting molecular properties, and proposed the neural message passing framework that unifies them. Xie, et al.(Crystal Graph Convolutional Neural Networks) studied graph neural networks to predict bulk properties of crystalline materials, and used a multi-graph named a crystal graph. …
read moreIn this work, we develop a new model (MT-CGCNN) by integrating CGCNN with transfer learning based on multi-task (MT) learning. We demonstrate the effectiveness of MT-CGCNN by simultaneous prediction of various material properties such as Formation Energy, Band Gap and Fermi Energy for a wide range of inorganic crystals (46774 …
read moreHere we present the deep learning architecture SchNet that is specifically designed to model atomistic systems by making use of continuous-filter convolutional layers. We demonstrate the capabilities of SchNet by accurately predicting a range of properties across chemical space for molecules and materials where our model learns chemically plausible embeddings …
read more