Dear Michael,
I noticed that you are testing Measuring four-point correlators, I am very interesting to calculate such correlation function too. Is it supported in the version(r47) and how to define this observable in the python file?
Best regards,
Jun-Jun Liang
Q: I want to define a correlator with operators defined on 4 different sites, such as , <O_k O_{k+1}="" O_l="" O_{l+1}="">. Can it be done in the MPS? (From Arya Dhar)
A:Measuring four-point correlators is not in the main version of OSMPS, but is a feature I added to a branch which is being tested by students in Lincoln's group. It can also be implemented directly in the Fortran code without too much effort. (Answered by Michael L. Wall)
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2015-03-31
Hi Jun-Jun,
The current version does not support general four point correlations. However, a special type of four point correlations can be done. It requires that four operators sit on contiguous sites. General four point correlation measurements will be available in the future.
Best
Wei
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Dear Wei,
Thank you. Then how such contiguous site correlation <O_kO_{k+1}O_lO_{l+1}> be defined in python file? It's difficult for me now to implement it directly in the Fortran code.
Best regards,
Jun-Jun Liang
Last edit: liang 2015-03-31
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
What I was referring to is not strictly four-point correlators, but rather
N-point terms can be added to an MPO. This is done with AddMPOTerm, using
'MBString' and passing ['O1','O2',...,'ON'] to the list of operators. This
will produce the term \sum_i O1_i O2_{i+1}... ON_{i+N} (note this
functionality does not enforce Hermiticity). In this way you could measure
a four-point correlator summed over all sites. We will be working on
implementing site-resolved four-point correlators in the near future.
Thanks.
Dear Wei,
Thank you. Then how such contiguous site correlation <O_k O_{k+1}="" O_l="" O_{l+1}=""> be defined in python file? It's difficult for me now to
implement it directly in the Fortran code.
Best regards,
Jun-Jun Liang
Dear Michael,
I noticed that you are testing Measuring four-point correlators, I am very interesting to calculate such correlation function too. Is it supported in the version(r47) and how to define this observable in the python file?
Best regards,
Jun-Jun Liang
Q: I want to define a correlator with operators defined on 4 different sites, such as , <O_k O_{k+1}="" O_l="" O_{l+1}="">. Can it be done in the MPS? (From Arya Dhar)
A:Measuring four-point correlators is not in the main version of OSMPS, but is a feature I added to a branch which is being tested by students in Lincoln's group. It can also be implemented directly in the Fortran code without too much effort. (Answered by Michael L. Wall)
Hi Jun-Jun,
The current version does not support general four point correlations. However, a special type of four point correlations can be done. It requires that four operators sit on contiguous sites. General four point correlation measurements will be available in the future.
Best
Wei
Dear Wei,
Thank you. Then how such contiguous site correlation <O_kO_{k+1}O_lO_{l+1}> be defined in python file? It's difficult for me now to implement it directly in the Fortran code.
Best regards,
Jun-Jun Liang
Last edit: liang 2015-03-31
Hi Jun-Jun,
What I was referring to is not strictly four-point correlators, but rather
N-point terms can be added to an MPO. This is done with AddMPOTerm, using
'MBString' and passing ['O1','O2',...,'ON'] to the list of operators. This
will produce the term \sum_i O1_i O2_{i+1}... ON_{i+N} (note this
functionality does not enforce Hermiticity). In this way you could measure
a four-point correlator summed over all sites. We will be working on
implementing site-resolved four-point correlators in the near future.
Thanks.
-Michael
On Tue, Mar 31, 2015 at 2:34 AM, liang liangjj929@users.sf.net wrote:
Dear Michael，
Thank you for your explanations that make me get more deep understanding about MPO operator rule.
Best regards
Jun-Jun