COMmon Bayesian Optimization Library (COMBO) is an open source python library for machine learning techniques. COMBO is amenable to large scale problems, because the computational time grows only linearly as the number of candidates increases. Hyperparameters of a prediction model can be automatically learned from data by maximizing type-II likelihood.
QMAS is an ab-initio electronic-structure computational code package based on the projector augmented-wave (PAW) with a plane wave basis set. It computes electronic states and various physical properties efficiently with high precision for a wide range of physical systems. It provides geometry optimization, electronic states in a static magnetic field, permittivity distribution at the atomic-scale, energy and stress distribution, positron annihilation parameters, and so forth.
Open source software for massively parallel quantum chemistry calculations. Energies and geometries of nano-sized molecules can be calculated without fragmentation. The program supports Hartree-Fock, density functional theory, and second-order Møller-Plesset perturbation theory calculations. The input format, execution method, and program structure are simple, and frequently used routines can be easily extracted.
MODYLAS is a highly parallelized general-purpose molecular dynamics (MD) simulation program appropriate for very large physical, chemical, and biological systems. It is equipped most standard MD techniques including free energy calculations based on thermodynamic integration method. Long-range forces are evaluated rigorously by the fast multipole method (FMM) without using the fast Fourier transform (FFT) in order to realize excellent scalability. The program enables investigations of large-scale real systems such as viruses, liposomes, assemblies of proteins and micelles, and polymers. It works on ordinary linux machines, too.
An open-source application for the first-principles calculation based on the all-electron method with localized bases. By adopting the full-potential LMTO method, high-speed electronic state calculation can be performed with a less number of bases compared with the standard all-electron method. There is no restriction on symmetries as in the LMTO-ASA method, and spin polarization and spin-orbit interaction can also be treated.
An application for semi-empirical quantum chemistry calculation. Special emphasis is placed on molecular dynamics simulations, and is able to run efficiently on large-scale cluster computer systems using OpenMP/MPI hybrid parallelism. The code is still under development, but the source code is distributed freely under the GPL license.
An open source application implementing path-integral Monte Carlo method based on Stochastic Green function method. Finite temperature calculation of extended Bose Hubbard model and Heisenberg model with finite field can be treated. JSON and YAML formats are adopted for data I/O.
An application for structure prediction based on the evolutionary algorithm. From an input of the atomic position in a unit cell and possible elements at each atomic position, this application predicts the stable structure and composition from the first-principles calculation and molecular dynamics in combination with the evolutionary algorithm. This application is written in Python, and uses Quantum ESPRESSO and GULP as an external program.
DDMRG (DynamicalDMRG) is a program for analyzing the dynamical properties of one-dimensional electron systems by using the density matrix renormalization group method. It simulates excited or photo-induced quantum phenomena in Mott insulators, spin-Peierls materials, organic materials, etc. Parallel computational procedures for linear and non-linear responses in low dimensional electron systems and analyzing routines for relaxation processes of excited states induced by photo-irradiation are available.
A sparse-modeling tool for computing the spectral function from the imaginary-time Green function. It removes statistical errors in quantum Monte Carlo data, and performs a stable analytical continuation. The obtained spectral function fulfills the non-negativity and the sum rule. The computation is fast and free from tuning parameters.