Dear Michael
I have tested the spinless fermi-hubbard model in the exmaple file and found that their groundstate energy agrees well with that from other method. So I constructed a common two-component Fermi-Hubbard according to spinless fermi-hubbard model, the related python language for building the model is in the following:
H=mps.MPO()
H.AddMPOTerm(Operators,'bond',['fdagger_0','f_0'],hparam='t',weight=-1.0,Phase=True)
H.AddMPOTerm(Operators,'bond',['fdagger_1','f_1'],hparam='t',weight=-1.0,Phase=True)
H.AddMPOTerm(Operators,'site','interaction',hparam='U',weight=1.0)
================
when run the python script, two problem occur.
1. an Exception is thrown: Operator interaction is not covariant.
then how I define interaction terms to satisfy "covariant".
2. when I remove interaction terms, the program give the groundstate energy but different from the value calculated with other methods. Then what's wrong with the model I defined?
The latex form of Hubbard model is :
H = -t \sum_{<i,j>} c^{\dagger}{i,\sigma} c{\sigma,j} +U\sum_{i}n_{i\uparrow}n_{i\downarrow}
Best regards
Jun-Jun Liang
Last edit: liang 2014-05-24
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
This code runs fine for me on the most recent version of the open source
code obtained from the repository (revision [32]). You can obtain that
code from the command line as
Dear Michael
I have tested the spinless fermi-hubbard model in the exmaple file and
found that their groundstate energy agrees well with that from other
method. So I constructed a common two-component Fermi-Hubbard according to
spinless fermi-hubbard model, the related python language for building the
model is in the following:
$H = -t \sum_{} c^{\dagger}{i,\sigma} c{\sigma,j}
+U\sum_{i}n_{i\uparrow}n_{i\downarrow} $ Build operators
H=mps.MPO()
H.AddMPOTerm(Operators,'bond',['fdagger_0','f_0']
,hparam='t',weight=-1.0,Phase=True)
H.AddMPOTerm(Operators,'bond',['fdagger_1','f_1']
,hparam='t',weight=-1.0,Phase=True)
H.AddMPOTerm(Operators,'site','interaction',hparam='U',weight=1.0)
================
when run the python script, two problem occur.
1. an Exception is thrown: Operator interaction is not covariant.
then how I define interaction terms to satisfy "covariant".
2. when I remove interaction terms, the program give the groundstate
energy but different from the value calculated with other methods. Then
what's wrong with the model I defined?
Dear Michael
I have tested the spinless fermi-hubbard model in the exmaple file and found that their groundstate energy agrees well with that from other method. So I constructed a common two-component Fermi-Hubbard according to spinless fermi-hubbard model, the related python language for building the model is in the following:
====================
Build operators
Operators=mps.BuildFermiOperators(spin=0.5)
Operators['interaction']=np.dot(Operators['nf_0'],Operators['nf_1'])
Define Hamiltonian MPO
H=mps.MPO()
H.AddMPOTerm(Operators,'bond',['fdagger_0','f_0'],hparam='t',weight=-1.0,Phase=True)
H.AddMPOTerm(Operators,'bond',['fdagger_1','f_1'],hparam='t',weight=-1.0,Phase=True)
H.AddMPOTerm(Operators,'site','interaction',hparam='U',weight=1.0)
================
when run the python script, two problem occur.
1. an Exception is thrown: Operator interaction is not covariant.
then how I define interaction terms to satisfy "covariant".
2. when I remove interaction terms, the program give the groundstate energy but different from the value calculated with other methods. Then what's wrong with the model I defined?
The latex form of Hubbard model is :
H = -t \sum_{<i,j>} c^{\dagger}{i,\sigma} c{\sigma,j} +U\sum_{i}n_{i\uparrow}n_{i\downarrow}
Best regards
Jun-Jun Liang
Last edit: liang 2014-05-24
Hi Jun-Jun Liang,
This code runs fine for me on the most recent version of the open source
code obtained from the repository (revision [32]). You can obtain that
code from the command line as
svn checkout svn://svn.code.sf.net/p/openmps/code/ openmps/
We fixed some issues with the quantum number conserving code when spins are
involved and this may have also fixed your problem. Thanks.
-Michael
On Sat, May 24, 2014 at 12:20 AM, liang liangjj929@users.sf.net wrote:
Michael,
Thank you, it works well.