4.2.2. CalcMod file

This file determines the parameters for the calculation method, model, and output mode. The file format is as follows.

CalcType   0
CalcModel   2
CalcEigenVec 0

File format

[string01] [int01]

Parameters

  • [string01]

    Type : String

    Description : Select a word from keywords.

  • [int01]

    Type : Int

    Description : A parameter that is correlated with a keyword.

Use rules

  • After setting the keywords at [string 01], a half-width blank is needed for setting a parameter.
  • Keywords can be set in random order.
  • If the keywords or filenames are incorrect, the program is terminated.
  • The keywords “CalcType” and “CalcModel” are essential.
  • When a head of line is “#”, the line is skipped.

Keywords and parameters

The parameters correlated with the keywords are as follows.

  • CalcType

    Type : Int

    Description : Select the method for calculation from the following list:
    0: Lanczos method
    1: Analysis of the physical properties by using TPQ
    2: Full diagonalization method.
    3: LOBCG for the ground state.

  • CalcModel

    Type : Int

    Description : Select the model from the following list:
    0: Fermion Hubbard model (canonical ensemble: conservation of particles or conservation of particles and the component of \(S_z\))
    1: Spin model (canonical ensemble: conservation of the component of \(S_z\))
    2: Kondo lattice model (canonical ensemble: conservation of particles, the component of \(S_z\))
    3: Fermion Hubbard model (grand canonical ensemble)
    4: Spin model (grand canonical ensemble)
    5: Kondo lattice model (grand canonical ensemble).

    For the fermion Hubbard model, you can select the model under the conservation of the particles by setting NCond in the ModPara file. When you want to select the model under the conservation of particles and the component of \(S_z\), set both NCond and 2Sz in the ModPara file.

  • CalcEigenVec

    Type : Int (default value: 0)

    Description : Select the method to calculate the eigenvectors:
    0: Lanczos+CG methods (when the convergence of eigenvectors is not sufficient for using the Lanczos method, the CG method is applied to calculate eigenvectors).
    1: Lanczos method.

  • InitialVecType

    Type : Int (default value: 0)

    Description : Select the type of an initial vector:
    0: Complex type
    1: Real type.

  • OutputEigenVec

    Type : Int (default value: 0)

    Description : Select the mode of outputting an eigenvector:
    0: Not output an eigenvector
    1: Output an eigenvector.

  • InputEigenVec

    Type : Int (default value: 0)

    Description : Select the mode of inputting an eigenvector:
    0: Not input an eigenvector
    1: Input an eigenvector.

  • ReStart

    Type : Int (default value: 0)

    Description : Select the mode of inputting a restart vector:
    0: Not restart calculation
    1: Output a restart vector
    2: Input a restart vector and output a new restart vector
    3: Input a restart vector.

  • CalcSpec

    Type : Int (default value: 0)

    Description : Select the mode of calculating dynamical Green’s functions:
    0: Not calculate dynamical Green’s functions
    1: (not restart) Input an initial vector and files for generating single excited or pair excited states
    2: Input components of triangular diagonal matrix
    3: Output both components of triangular diagonal matrix and a restart vector
    4: Input both components of triangular diagonal matrix and a restart vector
    5: Input and output both components of triangular diagonal matrix and a restart vector.

  • OutputHam

    Type : Int (default value: 0)

    Description : Full Diag)Select the mode of outputting Hamiltonian:
    0: not output Hamiltonian.
    1: output Hamiltonian.

  • InputHam

    Type : Int (default value: 0)

    Description : (Full Diag)Select the mode of inputting Hamiltonian:
    0: not input Hamiltonian.
    1: input Hamiltonian.

  • OutputExcitedVec

    Type : Int (default value: 0)

    Description : Select the mode of outputting an excited vector:
    0: Not output an eigenvector
    1: Output an eigenvector.

  • Scalapack

    Type : Int (default value: 0)

    Description : (Full Diag)Select to use ScaLAPACK library for full diagonalization:
    0: not to use ScaLAPACK.
    1: use ScaLAPACK.

  • NGPU

    Type : Int (default value: 2)

    Description : (Full Diag)Select the number of GPU devices for full diagonalization:
    \({\mathcal H} \Phi\) does not support to use GPU devices at multi-nodes.