Added a new choice (option = 2) to rescale the momentum after a hop: instead of ensuring that the total diagonal energy element is conserved, we ensure that total energy coming from the full adiabatic Hamiltonian matrix (including NAC) is conserved by using super adiabatic states.
Merge remote-tracking branch 'origin/master' into Leo_understanding_hop
obj <-> state typo
Added CodeMeta.json
After today's release of version 7.2.0
Happy New Year 2024!
Distinctly measuring CPU time and wall-clock time
Added new traj variable to ensure the continuity of the computed adiabatic eigenstates if needed
Merge remote-tracking branch 'origin/master'
Corrected super_adi according to the latest changes of eval_V_F
Draw textbox with trajectory details only at first time step
Cleaning up the sht variants
Running qm_setup|qm_cleanup before|after every simulation
Using default propagator, i.e. Strang-Marchuk
Running qm_setup|qm_cleanup before|after every simulation
Merge branch 'master' of https://git.code.sf.net/p/wavepacket/matlab/git
Using exactly the same time step counts for the four short time propagators
only calling solve_tdse_adi if the electronic amplitudes are really needed
Merge remote-tracking branch 'origin/master' into Leo_understanding_hop
Removed lvc_2D_type2
Using property 'empty'
Merge remote-tracking branch 'origin/master'
Merge remote-tracking branch 'origin/Leo_understanding_hop'
Getting ready for version 7.2.0 to be published "soon"
Fixing a stupid bug in efi/interp.m after 10 months
Added new time variable for saving the CPU and wall clock time as part of the global variable 'info'
Merge remote-tracking branch 'origin/Leo_understanding_hop'
Corrected the trapezoidal rule for propagating the electronic amplitudes. (Previous commits: fast-forward merge with master)
Using 'pot.generic' instead of 'pot.empty' throughout
Removed obsolete "global" statements for the "new" potential energy models
Changed the file names for the "new" potential energy models to lower case notation
Adding references for the "new" potential energy models
Model X was not running since obj.empty = false was missing
Added the trapezoidal rule to the propagation of the TDSE. Also renamed tdse_adi_2 to tdse_adi_euler to properly name the methods.
Modified the comments for calculation of the Hessian matrix of the potential energy
Some corrections regarding log -> prt
Added "obj.empty = false" to the new potential functions.
Merge remote-tracking branch 'origin/master' into Leo_understanding_hop
Added a new option of eq_motion which is a variant of QTSH
Added second derivative of the diabatic potential energies of Tully's systems.
Corrected a typo in the spin-boson method
WavePacket logo: Reduce canvas size
WavePacket logo: Now with color scheme from my web page, adapted from arXiv
Optionally localize pendular wave functions inside [-pi,pi] interval
Deleted initial Mises state because it is a special case of the initial pendular states
Added position and momentum shifts to exact pendular eigenstates
Added von Mises distribution as an initial wave function
Setting default values for initial Gaussians: pos_0 and width
Added multiplicity to cos-like dipole moment and polarizability
Fixed qm_bound for the full storage scheme
By MK: Made sparsification of matrix representation of kinetic operator much faster
By MK: Saving/loading kinetic operator from file for a series of qm_bound calculations
Fixed the version number in the README file
Not using complex-valued fields anymore
Simplified Trotter propagation
Simplified propagators for wave functions: Now without sub-step index k
Introducing a counter for the temporal sub-steps makes the dubious offset obsolete
Introducing a counter for the temporal sub-steps makes the dubious offset obsolete
Fixed error for very last sub-step in Strang propagation