Skip to content

CQMP/HubbardBenchmarks_Data

Repository files navigation

This is the main directory for the 2D Hubbard benchmark data.

This work is licensed under a [Creative Commons Attribution 4.0 International License](http://creativecommons.org/licenses/by/4.0/).

If you get stuck at any point please do not hesitate to contact egull@umich.edu or jamesleb@umich.edu (alternately jpfleblanc@gmail.com)

The data is in three main directories:

t1_tprime0.2 for t=1, t'=0.2
t1_tprime-0.2 for t=1, t'=-0.2
t1_tprime0 for t=1, t'=0

these directories contain folders with the name

tL_tprimeM_UN_nO_TP

with 
L the hopping parameter t
M the next-nearest hopping parameter t'
N the interaction strength U
O the density n
P the temperature T


Inside these directories are data from different methods, named after the observable and technique

E_technique.dat
D_technique.dat
M_techniqe.dat
Delta_technique.dat

where technique describes the technique used to generate the data. 

Technique List:
-------------
DMET: Density Matrix Embedding Theory data
DCA: Dynamical Cluster Approximation data
VMC: Fixed Node, Diffusion Monte Carlo
AFQMC: Auxilliary Field Quantum Monte Carlo
MRPHF: Multi-reference projected Hartree Fock
UCCSD: Unrestricted coupled cluster, singles and doubles
DF: Dual Fermions
DMRG: Density matrix renormalization group
DiagMC: Diagrammatic Monte Carlo
--------------

and O the observale, in particular:

E the Energy
D the Double Occupancy
M the magnetization 
Delta the superconducting order parameter

All data is given as:
n \Delta_n mu O \delta_O

where n is the density, \Delta_n the error on the density (zero for canonical methods, but non-zero for grand-canonical methods), O is the value of an observable, and \delta_O is the error on O, and mu is the chemical potential used to produce the given filling.  As a convention we will use is mu=0 to correspond to half filling, and ignore any Hartree shift in the chemical potential. 



In the case of results generated on finite systems, additional files should be present, so that extrapolations to the infinite system size can be reproduced. Filenames are:
E_technique_sizeLxL.dat, D_technique_sizeLxL.dat, etc.
or
E_technique_sizeN.dat, D_technique_sizeN.dat, etc.

-------------------
Diagrammatic Monte Carlo File descriptions
*************************************************
FILE NAME:   X_DiagMC_GGam
FORMAT:      n, err_n, mu, err_mu, X, err_X
EXPLANATION: results of G-Gamma series; simulated at fixed n; X= E or D

FILE NAME:   X_DiagMC_G0U-G0Gam0
FORMAT:      n, mu, err_mu, X, err_X
EXPLANATION: extrapolated by results of G0-U and G0-Gamma0 series simulated at fixed \mu

--------------------