COMPARED-COMPutation ARchive of Exact Diagonalization-

COMPARED

About

COMPutation ARchive of Exact Diagonalization (COMPARED) is a database(DB) for quantum lattice model simulations using the exact diagonalization method. Using HΦ, a quantum lattice model solver using the exact diagonalization method, and the supercomputer of the Institute for Solid State Physics (ISSP), exhaustive calculations are performed and the DB is created under a unified data standard. Scripts for the exhaustive calculation, DB generation, and a tool for obtaining information are also included in this repository. The calulation using the exact diagonalization method returns exact results under a specified lattice size and model, thus it is often used as verification data for comparison with other methods. We hope that this DB will be helpful for researchers and students in the field of condensed matter physics.

License

• Data base : CC BY 4.0
• Tools : GNU General Public License version 3

We hope that you cite the following URL and the reference when you publish the results using COMPARED:

• URL : https://isspns-gitlab.issp.u-tokyo.ac.jp/compared

• Reference : “Quantum lattice model solver HΦ”, M. Kawamura, K. Yoshimi, T. Misawa, Y. Yamaji, S. Todo, and N. Kawashima, Computer Physics Communications 217, 180 (2017).

Developers

• Kazuyoshi Yoshimi (Institute for Solid State Physics, The University of Tokyo)
• Yuichi Motoyama (Institute for Solid State Physics, The University of Tokyo)
• Kota Ido (Institute for Solid State Physics, The University of Tokyo)
• Ryo Tamura (International Center for Materials Nanoarchitectonics, National Institute for Materials Science)
• Yusuke Konishi (Academeia Co., Ltd.)

Usage

• Install DB

  • Using git

    git clone https://isspns-gitlab.issp.u-tokyo.ac.jp/compared/compared.git
    

  • Download directly

    wget [URL_to_download_data] -P [Path_to_store_download_data]
    

• How to get dataset from DB

  You can get dataset under some conditions from DB using serch_from_h5.py in tools directory by following procedures.

  1. Move to the directory where the DB file (*.h5) is located.

     • Note that some directories does not have the DB files because of the limitation of the storage. If you want to get the missing DBs, please contact us by e-mail (see the last part of this document).

  2. Make symbolic link to search_from_h5.py.

  3. Make an input file (hereafter, refferred to as search.in) for serch_from_h5.py.

  4. Enter the following command.

     python search_from_h5.py search.in
     

• How to generate DB using HΦ

  T.B.A.

Folder structure

• License

• README.md

• data

  • Spin

    • chain
      • l16
        • README.md
        • data_new.h5
        • stan.in
    • honeycomb
      • l2w2
        • README.md
        • stan.in
      • l2w3
        • README.md
        • data_new.h5
        • stan.in
      • l2w4
        • README.md
        • data_new.h5
        • stan.in
      • l3w3
        • README.md
        • data_new.h5
        • stan.in
    • kagome
      • l2w1
        • README.md
        • stan.in
      • l2w2
        • README.md
        • stan.in
      • l2w3
        • README.md
        • data_new.h5
        • stan.in
    • ladder
      • l8w2
        • README.md
        • data_new.h5
        • stan.in
    • square
      • l4w4
        • README.md
        • data_new.h5
        • stan.in
    • triangular
      • l6w3
        • README.md
        • data_new.h5
        • stan.in

  • Hubbard

    • chain
      • l10
        • README.md
        • data_new.h5
        • stan.in
      • l8
        • README.md
        • data_new.h5
        • stan.in
    • ladder
      • l4_w2
        • README.md
        • data_new.h5
        • stan.in
      • l6_w2
        • README.md
        • data_new.h5
        • stan.in

• tools

  • csv2h5.py
  • out_h5info.py
  • search_from_h5.py
  • tutorial_search_hdf5_en.md
  • util.py

Contact

e-mail: compared-dev__at__issp.u-tokyo.ac.jp (please replace __at__ by @)