Avoid a crash in plot_pads if no display can be found.
avoid overlapping subplots
On the cluster there is no display. PAD files are plotted and saved
simplify compilation of pylibxc wrapper. The include paths
Fragments: pad or truncate trajectories before analysing fragments
fix bug when initial_state = "brightest" or "fragment"
After removing the global import of matplotlib, the long header at
floats cannot be used as indices
Catch error when scratch folder is not writable and inform the user about it.
fix missing lapack library
include environment.yml in MANIFEST
Conda environment with all dependencies for the deprectated python 2.7
additional fragment is only created if there are unassigned atoms
FTDM is normalized by sum(P**2) and unassigned atom indices are added to a separate fragment
align CI coefficients before computing FTDM from wavevector, fix bug in interpolation of Slater-Koster tables
compute FTDM from electronic wavevector of trajectory, superposition of excited adiabatic states
backup
In COSMO the self-interaction of charge smeared out over a segment Sm is amm = 3.8 Sm^{-1/2}, added the square root
plot lone pairs in Lewis structures
boron with sextet in Lewis structures
show_molcoords.py can also compute Wiener number
read vibrational normal modes from molden file
revised QY paper
fixed bug in sline interpolation
singlet fission
Josef Michl's Festschrift
fixed a bug in the measurement of temperature, generate initial conditions for MD simulation from force field dynamics
backup
simple implementation of Free Energy Perturbation theory with dual topology
explained how command line arguments are parsed, read mol files in V3000 format
symmetrize molecule for given point group
write version string
manuscript for QY in J-aggregates, fixed reader for MOL format
read Lewis structure from MOL file
added changes from Freiburg/June/2020
atom types are assigned to each fragment separately
add random perturbation before optimizing 2D geometries
NAC vectors based on transition density matrix between excited states
simplex algorithm without matrix inversion
All Lewis structures are enumerated, but sometimes the matrix A_B is singular
All Lewis structures are enumerated, but sometimes the matrix A_B is singular
enumerate all resonance structures by brute force
constant T dynamics with DREIDING force field
compute FTDM on the fly, added example
fixed bug in dftbaby.sh
time series of FTDM for current state during dynamics
The fragments used for localizing orbitals can be specified as an option, allowing fragmentation of systems that do not consist of disconnected fragments
print warning if projection of initial diabatic onto adiabatic states is to low instead of aborting
visualize configurations from MD trajectories as 2D scatter plots using t-SNE
replaced OpenMP schedule(dynamic) with schedule(runtime)
show eigenvectors of eigenchannels, visualizer for molden files
limiting form for small x for Coulomb functions and spherical Bessel functions
project Dyson orbitals onto bound orbitals to remove component in the continuum
module for reading geometry, contracted spherical GTOs and MO coefficients from molden file
documentation of electroSka
identify resonances, read input from json file
read forces from Checkpoint files generated by QChem
fixed bug that led to beta2 outside valid interval [-1,2]
cleaned up code for (C)MS, there is still a bug in the calculation of PADs, beta > 2!
Sphinx compatible documentation of DFTB.MultipleScattering.MuffinTin
covalent radii for all elements
multiple scattering in muffin-tin approximation, eigenenergies of H2+ agree with MT+L=0 in Wastberg's paper
extract geometries from Gaussian log-file
NEB scan with Gaussian 09
NEB scan with Gaussian 09
backup
porphyrin Monte Carlo growth
repulsive potentials can be loaded from non-standard locations
molecular polyominos reactivity
plot reactivity of polyominos
allow to select solvent points based on lambda expression
removed OptionParserFuncWrapper from utils module
backup
bound orbitals can be normalized
backup
poster for Split
backup
kappa = 0 point is skipped
backup before cleaning up multiple scattering code
spurious solutions are eliminated based on the norm of the C^I coefficients
backup
backup
backup
something's still wrong
backup
backup
localized orbitals: overlap is computed between CSF's
backup
fixed bug in orbital localization
particle hole charges, orbital localization, initial guess for non-Hermitian Davidson diagonalization
backup
average fragment data
backup
fixed calculation of i_l(x) for x < 1
backup
E < V_II seems to work better, now
Now the same definition for k_l(x) is used as in Johnson's 1973 article
expansion theorems work, but I would like to remove the factor (-1)^l from the definition of k_l(x)
backup
upward iteration is not stable for modified spherical Bessel function of second kind i_l