- Comp-phys-alps-users - lists.phys.ethz.ch

3D quantum magnet
by Forwordom 07 Dec '12

07 Dec '12

Hi everyone I want to know if dirloop_sse QMC can simulation 3D finite size lattice such as magnetic clusters using Heisenberg model.

2 1

eigenvectors of T-matrix
by jessica.alfonsi＠unipd.it 06 Dec '12

06 Dec '12

Dear ALPS developers, is it possible to have the coeffcients and eigenvectors of the T-matrix returned from lanczos object in lanczos.h ? I need them (the first component to be exact) for computing the spectral function. I do see the member function t_eigenvectors() in the header, but I wouldn't know exactly how to write the code to get the values since I'm not much into C++. Could you please add a couple of lines of code later to one of the lanczos examples, so it can be helpful for all users ? Best regards, Jessica Alfonsi

1 0

Subject: Re: Sign problem
by Forwordom 06 Dec '12

06 Dec '12

Hi Matthias Thanks a lot for your reply. Yes, a nonzero EPSILON is only make the updates ergodic but can not eliminate sign problem. I cannot understand what your said "one can get rid of a sign problem by changing the basis". Actually, I have no any experience to modify or add codes in original ALPS codes. So I have to ask for you help. Now I send my lattice model and Hamiltonian model files to you. The used model is a coupled two ladder lattice with five exchange coupling constants in Heisenberg model Humiltonian.You can find those details in attachments. Two attach files can be found in attachments including my_ctwoladder_lattice describles lattice model and my_models describles the used model Humiltonian. Your reply will be highly appreciated. Forwordom

1 0

Sign problem
by Forwordom 05 Dec '12

05 Dec '12

Hi ALPS users and developers I meet sign problem in simulating frustrated system using dirloop_sse quantum Monte Carlo. As for this problem, the tutorial give some suggestions taht one should adjust parameter EPSILON and check the results carefully. But there is not the details of how to adjust this parameters. I have tried to modify all kinds of parameters including EPSILON,WHICH_LOOP_TYPE and so on. However the warning "the hamiltonian has sign problem" always appears when running. Can I get reliable the results when sign problem exists? How to modify simulation parameters to avoid sign problem using dirloop_sse method for my simulating frustrated systems. Can you give help. Thanks advanced.

2 1

question
by sh po 04 Dec '12

04 Dec '12

Hi friends.Recently I installed Alps on my ubuntu linux . sorry but I read your site carefully but it was so difficult for me to understand the way of working with alps.Because Im not professional enough in linux. I have 2 question from you 1.I was wondering if you'd mind giving me a text file (for example pdf) which explains way of working with alps in an easy and complete way. 2.My other question is that where should I type the codes that you wrote in your site? I mean should I copy and past them in my terminal? or copy and past them in a text filr which I open with "gedit" in linux?If a text file, so with what type I must save that text file?and after that with what compiler I must run it on my terminal? thanks alot for your help.

2 2

Local observables in quantum montecarlo
by Francisco Cordobés 29 Nov '12

29 Nov '12

Dear all, I am trying to understand how to use the qmc codes from ALPS to get some local observables from lattice models. If I understand it correctly, the way qmc codes work is in imaginary time and in that representation one can measure thermodynamic quantities. In the case of dynamical quantities(by this I mean dynamical susceptibilities), the results in imaginary time have to be analytically continued back to the real time axis and then evaluated. This is done usually, from what I gather from the literature, using maximum entropy methods. I have read through the tutorials on the alps website and the different slides by the ALPS creators that I found on google and I haven't found any example of the last type of calculations. Neither how to "tell" ALPS to measure a non-averaged susceptibility or how to analytically continue it to the real axis. Is it possible with ALPS as it is or do I need to write my own c++ code, based on ALPS, to access that kind of information? If it is possible, could someone tell me where I can find documentation on how to do it specifically with ALPS? Kind regards, Paco Cordobes

1 0

Fwd: honeycomb bilayer
by mojtaba alaei 29 Nov '12

29 Nov '12

Dear All Recently I tried to reproduce the result of a paper about two honeycomb bilayer (PRB.85.184412 <http://prb.aps.org/forward/PRB/v85/i18/e184412>) . The paper used simple MC simulations ( in fact authors used exchange MC technique <http://dx.doi.org/10.1143/JPSJ.65.1604>) with classical Heisenberg model to derive susceptibility of this system. Authors used three different exchange parameters: 1-J1 as NN interaction in honeycomb surfaces 2-J2 as NNN interactions in honeycomb surfaces 3-Jc as face-to face intra-bilayer interactions. I used exactly the same parameters of the paper (I use minus sign for exchange). My setup is as follows: ****************************************** ****************************************** Exchange parameters setting: J1=30.2 K, J2/J1=0.1 and Jc/J1=0.1. Landee g-factor: g=2 steps for thermalization: 50000 steps for sampling: 1000000 lattice size: L=24 ****************************************** ****************************************** All of my parameters are the same as the paper except that I used more steps for sampling and for thermalization because the system is frustrated. The following plots are mine and from the paper ( I derived the data of fig(8,c) of the paper by using G3DATA): The ALPS result (mine) [image: Inline image 1] The PRB.85.184412 result (please see fig 8_c of the paper): [image: Inline image 2] *As you can see the plot of mine has a maximum around 40 K while the plot of PRB.85.184412 has a maximum around 60-70 K and also the plot shows a minimum around 20 K. W**hat do you think? what is my fault? Is it a bug or something else?* If somebody like to repeat the calculation, these are my input files: the python script: #================================= import pyalps import matplotlib.pyplot as plt import pyalps.plot #prepare the input parameters parms = [] for t in [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 22, 24, 26, 28, 30, 32, 34, 36, 38, 40, 43, 46, 49, 52, 55, 58, 61, 64, 67, 70, 75, 80, 85, 90, 95, 100, 110, 120, 130, 140, 150, 160, 170, 180, 190, 200, 210, 220, 230, 240, 250, 260, 270, 280, 290, 300, 310, 320, 330, 340, 350, 360, 370, 380, 390, 400 ]: parms.append( { 'LATTICE_LIBRARY' : "graphene_bilayer.xml", 'LATTICE' : "graphene bilayer lattice", 'S' : 1.5, 'T' : t, 'J1' : -30.2, 'J2' : -3.02, 'J3' : -3.02, 'g' : 2.00, 'THERMALIZATION' : 50000, 'SWEEPS' : 1000000, 'UPDATE' : "local", 'MODEL' : "Heisenberg", 'L' : 24 } ) #write the input file and run the simulation input_file = pyalps.writeInputFiles('parm2b',parms) pyalps.runApplication('spinmc',input_file,Tmin=300, MPI=4, mpirun='mpirun') #get the list of result files result_files = pyalps.getResultFiles(prefix='parm2b') print "Loading results from the files: ", result_files #print the observables stored in those files: print "The files contain the following mesurements:", print pyalps.loadObservableList(result_files) #load a selection of measurements: data = pyalps.loadMeasurements(result_files,['Susceptibility']) #make a plot for the magnetization: collect Magnetziation as function of T plotdata = pyalps.collectXY(data,'T','Susceptibility') # convert the data to text file for plotting using another tool print "The results in plain text are:" print pyalps.plot.convertToText(plotdata) print "Saving into file parm2b.txt" f = open ('parm2b.txt','w') f.write(pyalps.plot.convertToText(plotdata)) f.close() #================================ the lattice file (xml format) ===================================== <LATTICES> <LATTICE name="hexagonal lattice" dimension="3"> <PARAMETER name="a" default="1"/> <PARAMETER name="c" default="1"/> <BASIS> <VECTOR>a 0 0</VECTOR> <VECTOR>a/2 a*sqrt(3)/2 0</VECTOR> <VECTOR>0 0 c</VECTOR> </BASIS> <RECIPROCALBASIS> <VECTOR>2*pi/a -2*pi/a/sqrt(3) 0 </VECTOR> <VECTOR>0 4*pi/a/sqrt(3) 0 </VECTOR> <VECTOR>0 0 2*pi/c</VECTOR> </RECIPROCALBASIS> </LATTICE> <UNITCELL name="graphene bilayer" dimension="3" vertices="4"> <VERTEX><COORDINATE> 0.000000000000 0.0000000000 0 </COORDINATE></VERTEX> <VERTEX><COORDINATE> 0.333333333333 0.3333333333 0</COORDINATE></VERTEX> <VERTEX><COORDINATE> 0.000000000000 0.0000000000 0.5 </COORDINATE></VERTEX> <VERTEX><COORDINATE> 0.333333333333 0.3333333333 0.5</COORDINATE></VERTEX> <EDGE type="1"><SOURCE vertex="1"/><TARGET vertex="2" offset=" 0 0 0"/></EDGE> <EDGE type="1"><SOURCE vertex="3"/><TARGET vertex="4" offset=" 0 0 0"/></EDGE> <EDGE type="1"><SOURCE vertex="2"/><TARGET vertex="1" offset=" 1 0 0"/></EDGE> <EDGE type="1"><SOURCE vertex="2"/><TARGET vertex="1" offset=" 0 1 0"/></EDGE> <EDGE type="1"><SOURCE vertex="4"/><TARGET vertex="3" offset=" 1 0 0"/></EDGE> <EDGE type="1"><SOURCE vertex="4"/><TARGET vertex="3" offset=" 0 1 0"/></EDGE> <EDGE type="2"><SOURCE vertex="1"/><TARGET vertex="3" offset=" 0 0 0"/></EDGE> <EDGE type="2"><SOURCE vertex="2"/><TARGET vertex="4" offset=" 0 0 0"/></EDGE> <EDGE type="3"><SOURCE vertex="1"/><TARGET vertex="1" offset=" 1 0 0"/></EDGE> <EDGE type="3"><SOURCE vertex="1"/><TARGET vertex="1" offset=" 0 1 0"/></EDGE> <EDGE type="3"><SOURCE vertex="1"/><TARGET vertex="1" offset=" 1 -1 0"/></EDGE> <EDGE type="3"><SOURCE vertex="2"/><TARGET vertex="2" offset=" 1 0 0"/></EDGE> <EDGE type="3"><SOURCE vertex="2"/><TARGET vertex="2" offset=" 0 1 0"/></EDGE> <EDGE type="3"><SOURCE vertex="2"/><TARGET vertex="2" offset=" 1 -1 0"/></EDGE> <EDGE type="3"><SOURCE vertex="3"/><TARGET vertex="3" offset=" 1 0 0"/></EDGE> <EDGE type="3"><SOURCE vertex="3"/><TARGET vertex="3" offset=" 0 1 0"/></EDGE> <EDGE type="3"><SOURCE vertex="3"/><TARGET vertex="3" offset=" 1 -1 0"/></EDGE> <EDGE type="3"><SOURCE vertex="4"/><TARGET vertex="4" offset=" 1 0 0"/></EDGE> <EDGE type="3"><SOURCE vertex="4"/><TARGET vertex="4" offset=" 0 1 0"/></EDGE> <EDGE type="3"><SOURCE vertex="4"/><TARGET vertex="4" offset=" 1 -1 0"/></EDGE> </UNITCELL> <LATTICEGRAPH name = "graphene bilayer lattice"> <FINITELATTICE> <LATTICE ref="hexagonal lattice"/> <PARAMETER name="W" default="L"/> <PARAMETER name="H" default="1"/> <EXTENT dimension="1" size="L"/> <EXTENT dimension="2" size="W"/> <EXTENT dimension="3" size="H"/> <BOUNDARY type="periodic"/> </FINITELATTICE> <UNITCELL ref="graphene bilayer"/> </LATTICEGRAPH> </LATTICES> ======================================================================= Thanks in advanced Mojtaba Alaei

1 0

non-next-neighbor Hamiltonians
by Emanuele Levi 27 Nov '12

27 Nov '12

Dear ALPS developers, does ALPS ( and in particular the DMRG solver) support spin Hamiltonians with non-next interacting terms, e.g. Sz(i)Sz(i+3)? Many thanks. Kindest regards. -- Emanuele Levi emanuele.levi(a)gmail.com

2 6

Resuming a DMRG simulation
by A. Giovanni Ramirez Garcia 23 Nov '12

23 Nov '12

Dear All, I'm using the DMRG code in the alps-2.1.0d1-r6050 version and I would like to resume some simulations after they have finish properly. I want to increase the number of sweeps previously defined (sometimes from 3 to 4) then when it finishes I just add the parameters to my python script to start from the sweep 3. When I ran the script it stops due an error of *.dat files not found. Where ALPS save the *.dat files to resume from another simulation? I tried to find them by myself without success. Should I give another parameter to let ALPS knows I want to save that files? Giovanni Ramirez-Garcia

1 0

how to retrieve output datas in DMRG
by Emanuele Levi 19 Nov '12

19 Nov '12

Dear ALPS users, I am running the DMRG algorithm with python. I don't understand where the energy/entropy/trunc. err. history is recorded. I would like to set up a way to create three different files with the history of these three quantities. Is there a way to do that? Many thanks. Kindest regards. -- Emanuele Levi emanuele.levi(a)gmail.com

2 1