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

Bls: Comp-phys-alps-users Digest, Vol 40, Issue 3
by Muhammad Rofiq 03 Sep '09

03 Sep '09

my name is Rofiq. I have a problem for the plotting magnetization graph of my reaserch... So, i would to ask" can The ALPS Used to plott magetization and temperature on a graph?" Yahoo! Toolbar kini dilengkapi Anti-Virus dan Anti-Adware gratis. Download Yahoo! Toolbar sekarang. http://id.toolbar.yahoo.com

1 0

biquadratic terms in QMC
by Kazuo Hida 31 Aug '09

31 Aug '09

Dear all Thank you for developping ALPS. I am just trying to use ALPS QMC library. I have a simple question. Does any of ALPS QMC code for quantum spin models support the biquadratic terms (\vec{S}_i\cdot\vec{S}_j)^2 for S \geq 1? With best regards Kazuo Hida Division of Material Science Graduate School of Science and Engineering Saitama University Saitama, 338-8570 JAPAN Tel&FAX:+48-858-3613 e-mail: hida(a)phy.saitama-u.ac.jp Homepage: http://www.phy.saitama-u.ac.jp/~hida/index_e.html

4 4

Re: [ALPS-users] inverse-sqare 1d ising chain (Synge Todo)
by Tadeusz Wasiutynski 24 Aug '09

24 Aug '09

Dear, Is there a method for the simulation of the system with competing interactions e.g dipolar r^{-3} and quadrupolar r^{-5} distance dependence? From the paper Fukui and Todo it is not clear for me. Best regards, Tadeusz On Sat, 22 Aug 2009 12:00:03 +0200, comp-phys-alps-users-request wrote > > Message: 1 > Date: Fri, 21 Aug 2009 19:07:39 +0900 > From: Synge Todo <wistaria(a)comp-phys.org> > Subject: Re: [ALPS-users] inverse-sqare 1d ising chain > To: Andrea Taroni <andrea.taroni(a)fysik.uu.se>, > comp-phys-alps-users(a)phys.ethz.ch > Message-ID: <EDEE50B2-E084-4602-B562-AA97D75E62B3(a)comp-phys.org> > Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes > > Dear Andrea, > > Currently, ALPS lattice and model libraries do not support long- > range interactions, though you can simulate such models by > specifying a graph with N(N-1)/ 2 bonds with different bond types > for different bond lengths (and it will take O(N^2) CPU time). > > Or, I can send you our QMC code used in Fukui and Todo (2009) if you > want. > > Best, > Synge > > On 2009/08/21, at 17:45, Andrea Taroni wrote: > > > Dear All, > > > > Is there a straightforward way to simulate the classical inverse- > > square > > Ising model using ALPS? By straightforward, I mean using the kind of > > recipe that is described in the tutorials section of the APLS > > applications website. > > > > I am interested in comparing any eventual results with those presented > > by Fukui and Todo J. Comp. Phys. 228, 2629 (2009), using a novel > > order-N > > cluster MC method. > > > > Thanks in advance! > > > > Andrea > > End of Comp-phys-alps-users Digest, Vol 41, Issue 3 > *************************************************** Prof. dr hab. Tadeusz Wasiutynski H. Niewodniczanski Institute of Nuclear Physics PAN tel. +48(12)6628251 fax: +48(12)6628251 e-mail: tadeusz.wasiutynski(a)ifj.edu.pl

2 1

inverse-sqare 1d ising chain
by Andrea Taroni 21 Aug '09

21 Aug '09

Dear All, Is there a straightforward way to simulate the classical inverse-square Ising model using ALPS? By straightforward, I mean using the kind of recipe that is described in the tutorials section of the APLS applications website. I am interested in comparing any eventual results with those presented by Fukui and Todo J. Comp. Phys. 228, 2629 (2009), using a novel order-N cluster MC method. Thanks in advance! Andrea

2 1

only some disorder seeds working in sparsediag?
by Bob Gooding 20 Aug '09

20 Aug '09

There may be a problem with sparsediag (my previous coding error notwithstanding). Here is the parameter file used for both sparsediag and fulldiag LATTICE_LIBRARY = "/usr/local/ALPS/lib/xml/lattices.xml"; LATTICE = "inhomogeneous simple cubic lattice"; MODEL_LIBRARY= "/usr/local/ALPS/lib/xml/models.xml"; MODEL = "alternative fermion Hubbard"; CONSERVED_QUANTUMNUMBERS="N,Sz"; MEASURE_LOCAL[Local charge density]=n; t = 1.0; L = 2; U = 9.0; N_total = 4; Sz_total = 0; TRANSLATION_SYMMETRY = false; {DISORDERSEED = 11138; mu = -6.0*(random()-0.5);} Here is the relevant portion of the task1.out file generated by sparsediag - the densities are all zero, but the ground state energy is correct (see below). <EIGENVALUES number="1"> -11.43877071841712 </EIGENVALUES> <EIGENSTATES number="1"> <EIGENSTATE number="0"> <SCALAR_AVERAGE name="Energy"><MEAN>-11.43877071841712</MEAN></SCALAR_AVERAGE> <VECTOR_AVERAGE name="Local charge density"> <SCALAR_AVERAGE indexvalue="( 0,0,0 )"><MEAN>0</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 0,0,1 )"><MEAN>0</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 0,1,0 )"><MEAN>0</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 0,1,1 )"><MEAN>0</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 1,0,0 )"><MEAN>0</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 1,0,1 )"><MEAN>0</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 1,1,0 )"><MEAN>0</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 1,1,1 )"><MEAN>0</MEAN></SCALAR_AVERAGE> </VECTOR_AVERAGE> Here is a portion of the output from fulldiag - same energy as sparsediag, but now the local charge densities are non zero. <EIGENVALUES number="784"> -11.43877071841714 </EIGENVALUES> ... <EIGENSTATE number="0"> <SCALAR_AVERAGE name="Energy"><MEAN>-11.43877071841714</MEAN></SCALAR_AVERAGE> <VECTOR_AVERAGE name="Local charge density"> <SCALAR_AVERAGE indexvalue="( 0,0,0 )"><MEAN>0.3727847497057824</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 0,0,1 )"><MEAN>0.8699453754269064</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 0,1,0 )"><MEAN>0.4528964351309576</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 0,1,1 )"><MEAN>0.7330138222807983</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 1,0,0 )"><MEAN>0.6288755119873222</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 1,0,1 )"><MEAN>0.1885341339328754</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 1,1,0 )"><MEAN>0.3029849265253151</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 1,1,1 )"><MEAN>0.4509650450100266</MEAN></SCALAR_AVERAGE> </VECTOR_AVERAGE>

1 0

only some disorder seeds working in sparsediag?
by Bob Gooding 19 Aug '09

19 Aug '09

If I use mu = -W*(rand()-1/2) then this is equivalent in to the Anderson Hubbard model with random on-site energies - a box distribution of strength W. (That is, the grand Hamiltonian H - mu*N becomes the Anderson Hubbard Hamiltonian (H only).) We are completing our work in the fixed particle number canonical ensemble, and my understanding of the conserved quantum numbers option is that if I specify N_total=4 (say) then the calculation is completed in the canonical ensemble. So, perhaps my error (???) is in the use of CONSERVED_QUANTUMNUMBERS="N_total"; N_total=4; > Could it be just that the values of mu are such that the ground state has no particles? > Matthias

2 1

only some disorder seeds working in sparsediag?
by Bob Gooding 19 Aug '09

19 Aug '09

We are using some of our own lattices and models, but the problem we are encountering can be identified using existing lattices ("inhomogeneous square lattice" and "inhomogeneous simple cubic lattice") and different fermion Hubbard models. As one concrete example, with the following (simplified) parameter file, only for some of the runs does one get local density data. Two example outputs are shown. Note that the same thing occurs (of course) if we use two separate parameter files, each one listing a particular disorder seed. Seeds 11113, 11114 work, 11115 does not, etc. Also, the problem exists for N_total=6 and 4 ... So, do constraints exist on valid choices of DISORDERSEED? (Also, since we would like to know the complexion of disorder associated with a given ground state wave function, we would in fact like to know the random number generator being used.) Thanks. ---- LATTICE_LIBRARY = "/usr/local/ALPS/lib/xml/lattices.xml"; LATTICE = "inhomogeneous simple cubic lattice"; MODEL_LIBRARY= "/usr/local/ALPS/lib/xml/models.xml"; MODEL = "alternative fermion Hubbard"; CONSERVED_QUANTUMNUMBERS="Sz_total"; CONSERVED_QUANTUMNUMBERS="N_total"; MEASURE_LOCAL[Local charge density]=n; t = 1.0; U = 2.0; L = 2; Sz_total = 0; N_total = 8; TRANSLATION_SYMMETRY = false; { DISORDERSEED = 11111; mu=-6.0*(random()-0.5); } { DISORDERSEED = 11112; mu=-6.0*(random()-0.5); } some of task1 output: <EIGENVALUES number="1"> -13.57576898386332 </EIGENVALUES> <EIGENSTATES number="1"> <EIGENSTATE number="0"> <SCALAR_AVERAGE name="Energy"><MEAN>-13.57576898386332</MEAN></SCALAR_AVERAGE> <VECTOR_AVERAGE name="Local charge density"> <SCALAR_AVERAGE indexvalue="( 0,0,0 )"><MEAN>1.02945417967653</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 0,0,1 )"><MEAN>0.9407378693270886</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 0,1,0 )"><MEAN>0.8692266888738961</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 0,1,1 )"><MEAN>0.2243604886901401</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 1,0,0 )"><MEAN>0.9836579726343326</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 1,0,1 )"><MEAN>0.9533153322138351</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 1,1,0 )"><MEAN>1.256609332250588</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 1,1,1 )"><MEAN>1.742638136334122</MEAN></SCALAR_AVERAGE> </VECTOR_AVERAGE> </EIGENSTATE> </EIGENSTATES> some of task2 output: <EIGENVALUES number="1"> -12.3429442715644 </EIGENVALUES> <EIGENSTATES number="1"> <EIGENSTATE number="0"> <SCALAR_AVERAGE name="Energy"><MEAN>-12.3429442715644</MEAN></SCALAR_AVERAGE> <VECTOR_AVERAGE name="Local charge density"> <SCALAR_AVERAGE indexvalue="( 0,0,0 )"><MEAN>0</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 0,0,1 )"><MEAN>0</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 0,1,0 )"><MEAN>0</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 0,1,1 )"><MEAN>0</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 1,0,0 )"><MEAN>0</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 1,0,1 )"><MEAN>0</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 1,1,0 )"><MEAN>0</MEAN></SCALAR_AVERAGE> <SCALAR_AVERAGE indexvalue="( 1,1,1 )"><MEAN>0</MEAN></SCALAR_AVERAGE> </VECTOR_AVERAGE> </EIGENSTATE> </EIGENSTATES>

2 1

Local Minima in DMRG
by Jesse Kinder 16 Jul '09

16 Jul '09

I was looking into the convergence of the ground state energy in DMRG as a function of NUMSTATES. For instance, on a 1D chain of spinless fermions with open boundary conditions at half-filling (L=36, N_TOTAL=12) and interaction strength V0=4, the ground state energies after 30 sweeps are M=2 E0 = -8.6661 M=4 E0 = -9.6249 M=8 E0 = -9.7115 M=100 E0 = -9.7133 The truncation error for the M=2 run was just 0.00167, so it is surprising (to me, at lest) that the energy is off by more than 10 percent. A discrepancy of 10 percent or more occurs for chains of different lengths and interaction potentials of different strengths. Doing 100 sweeps instead of 30 has no effect on the first 5 digits of the M=2 results. Nor does increasing NUM_WARMUP_STATES to 100. One possibility is that the calculation is getting stuck in a local minimum. Does the ALPS DMRG routine have any method for avoiding local minima? For instance, adding a small amount of noise to the calculation to allow tunneling out of local minima? Even if the results of the above calculations simply reflect the poor quality of the M=2 wave function, it would still be useful to know how to handle local minima that might occur in DMRG calculations. Jesse

2 1

Green Fuction and Local Magnetization with looper
by João Manuel Viana Parente Lopes 07 Jul '09

07 Jul '09

Dear Alps list, Greatings from Angola... I am trying to extract the Green Function and the local magnetization running the loop code. I tried into the simplest example.. param file: LATTICE="open chain lattice" MODEL="spin" local_S=1/2 L=10 J=1 INITIAL_SITE=1 THERMALIZATION=15000 SWEEPS=150000 REPRESENTATION="SSE" MEASURE[Green Function]=true MEASURE[Correlations]=true MEASURE[Local Susceptibility]=true {T=0.05;} the simulation runs in 1m but there is no output for either Green's Funtion or Local Magnetization... Am I doing something terrible wrong or the looper code is not given these measures? Thanks in Advance João

2 1

complex hopping parameter
by Jakub Kolasinski 19 Jun '09

19 Jun '09

Dear Mr Troyer, I allow myself to use Your kindness one more time. My question is, how can I make hopping parameter complex (if that is possible at all)? I get an error that "can not convert complex number into a real one". I'm using worm. Thank You in advance for Your answer Jakub

2 1