kpm

Submodules:

• kpm.svect
• kpm.disc
• kpm.mom
• kpm.kern
• kpm.rcstr
• kpm.resc
• kpm.plot
• kpm.cf
• kpm.misc

Implement the kernel polynomial method (KPM) [1].

Right now, KPM algorithms for the following target quantities are provided:

• Density of states (DOS)
• Local density of states (LDOS)

By averaging (arithmetic or geometric mean), also the following quantities can be calculated:

• Arithmetic mean of the LDOS (ALDOS)
• Geometric mean of the LDOS (GLDOS)

The algorithms expect either tight binding matrices, or supercell definitions as defined in the module sc which provide rules to create a matrix “on the fly”.

Certain submodules are written in Cython [2] to obtain better performance and allow for OpenMP parallelization.

References:

• [1] Weiße et al., Rev. Mod. Phys. 78, 275 (2006)
• [2] http://cython.org/

Functions

kpm.ldos(mat, state=0, limit=100, erange=10, enum=None, estep=None, kernel='jackson', rcstr_method='std', omp=False, num_threads=None, rescaled=False, stateclass='')

Calculate local density of states (LDOS) from the given tight binding matrix. Return energy and density array.

kpm.ldos_vm(mat, varmom, state=0, erange=10, enum=None, estep=None, kernel='jackson', rcstr_method='std', omp=False, num_threads=None, rescaled=False, stateclass='')

Calculate local density of states (LDOS) using the given tight binding matrix mat and the energy-dependent number of Chebychev moments varmom. Return energy and density array.

kpm.aldos(scell, erange=10, enum=None, estep=None, count=None, tol=None, smooth=10, limit=100, kernel='jackson', rcstr_method='std', stateclass=None, spr=1, omp=False, num_threads=0, until=None, verbose=False, init_dens=None, init_var=None, init_count=None)

Calculate arithmetic mean of local density of states (ALDOS).

kpm.gldos(scell, erange=10, enum=None, estep=None, count=None, tol=None, smooth=10, limit=100, kernel='jackson', rcstr_method='std', stateclass=None, spr=1, omp=False, num_threads=0, until=None, verbose=False, init_dens=None, init_var=None, init_count=None)

Calculate geometric mean of local density of states (GLDOS) (also known as the typical density of states).

kpm.galdos(scell, erange=10, enum=None, estep=None, count=None, tol=None, smooth=10, limit=100, kernel='jackson', rcstr_method='std', stateclass=None, spr=1, omp=False, num_threads=0, until=None, verbose=False, init_adens=None, init_avar=None, init_acount=None, init_gdens=None, init_gvar=None, init_gcount=None)

Calculate both the geometric (typical average) and the arithmetic mean of the local density of states (GLDOS and ALDOS) at the same time, using each local density twice. In this way, the numerical effort is easily reduced by a factor of 2 if both types of averages are needed.

kpm.dos(mat, rcount=None, rtol=None, rsmooth=2, limit=100, erange=10, enum=None, estep=None, kernel='jackson', rcstr_method='std', omp=False, num_threads=None, rescaled=False, until=None, verbose=False)

Calculate density of states (DOS) using “stochastic evaluation of traces”.

Note that there is no ensemble averaging. Use the function ados to include an average over different disorder configurations if a random system is studied.

kpm.ados(scell, erange=10.0, enum=None, estep=None, count=None, tol=None, smooth=10, rcount=None, rtol=None, rsmooth=2, limit=100, kernel='jackson', rcstr_method='std', omp=False, num_threads=0, until=None, verbose=False, init_dens=None, init_var=None, init_count=None)

Calculate arithmetic mean of density of states (ADOS) (ensemble average).

kpm.glcount(filename)

Get the attribute gldos.attrs.count of the dataset gldos from the given HDF5 file.

kpm.alcount(filename)

Get the attribute aldos.attrs.count of the dataset aldos from the given HDF5 file.

kpm.glacc(filename)

Get the attribute gldos.attrs.acc of the dataset gldos from the given HDF5 file.

kpm.alacc(filename)

Get the attribute aldos.attrs.acc of the dataset aldos from the given HDF5 file.

kpm.gllimit(filename)

Get the attribute gldos.attrs.limit of the dataset gldos from the given HDF5 file.

kpm.allimit(filename)

Get the attribute aldos.attrs.limit of the dataset aldos from the given HDF5 file.

kpm.acount(filename)

Get the attribute ados.attrs.count of the dataset ados from the given HDF5 file.

kpm.aacc(filename)

Get the attribute ados.attrs.acc of the dataset ados from the given HDF5 file.

kpm.alimit(filename)

Get the attribute ados.attrs.limit of the dataset ados from the given HDF5 file.

kpm.lcount(filename)

Get the attribute ldos.attrs.count of the dataset ldos from the given HDF5 file.

kpm.lacc(filename)

Get the attribute ldos.attrs.acc of the dataset ldos from the given HDF5 file.

kpm.llimit(filename)

Get the attribute ldos.attrs.limit of the dataset ldos from the given HDF5 file.

kpm.dcount(filename)

Get the attribute dos.attrs.count of the dataset dos from the given HDF5 file.

kpm.dacc(filename)

Get the attribute dos.attrs.acc of the dataset dos from the given HDF5 file.

kpm.dlimit(filename)

Get the attribute dos.attrs.limit of the dataset dos from the given HDF5 file.

kpm.glstderr0(filename)

Return the standard error of the GLDOS at zero energy.

kpm.alstderr0(filename)

Return the standard error of the ALDOS at zero energy.

kpm.astderr0(filename)

Return the standard error of the ADOS at zero energy.

kpm.checksigma(gldos_list, aldos_list)

Investigate the fluctuation of the standard error of the geometric mean among independent calculations.

kpm.glval0(filename)

Return the GLDOS at zero energy.

kpm.alval0(filename)

Return the ALDOS at zero energy.

kpm.aval0(filename)

Return the ADOS at zero energy.

kpm._get_stateclass(stateclass_option, stateclasses)

Return state indices according to the selected stateclass(es).

kpm._get_energ(rcstr_func, enum, estep, params)

kpm._get_rcstr_args(rcstr_func, energ, enum, omp, num_threads)

Indices and tables

• Index
• Module Index
• Search Page