July 2017 - Comp-phys-alps-users

measuring truncation errors in mps_optim
by Robertson Esperanza 07 Dec '17

07 Dec '17

Thank you for all the help so far, now I'm trying to check for convergence of my results. First, is it enough to just check how the truncation error changes with sweeps (at a certain MAXSTATE value) or I need to both check the iteration history of the ground state energy and the truncation error? Second, checking the supplementary materials provided (Computer Physics Communications 185 (2014) 3430–3440) for the Hubbard ladder, why are there 381 values of energy and truncation weights for every iteration? Why is the total truncation error for that iteration is the sum of all the 381 truncation weight values (based on iteration.py)? How is truncation error defined (e.g. is it 1 - sum of all reduced density eigenvalues corresponding to all the picked states?) ? As far as I understand the DMRG algorithm every iteration/sweep has a corresponding bond dimension in which as you progress through iterations (until you reach the "SWEEPS"th iteration), bond dimension increases until you reach "MAXSTATES" at the "SWEEPS"th iteration. Regards, Robertson Esperanza

4 10

Problem on finding xsl path
by Jin Hyoun Kang 19 Aug '17

19 Aug '17

2 1

Two-dimensional system with MPS
by H 13 Jul '17

13 Jul '17

Dear Michele or others, In your paper (arXiv:1407.0872)'s abstract you mentioned the MPS code can handle 2D system. Is it as simple as setting the 'LATTICE' parameter to "open square lattice", or do I need to build a custom 1D lattice with long-range coupling? I tried setting 'LATTICE' to "open square lattice" but the results seems different from my ED result. The system is spinful fermionic. Could it be due to fermion signs not properly taken care of internally? Ultimately I need to simulate a small 3D system (3x3x3). Can I use the internal cubic lattice or do I need to build a custom 1D lattice with long-range coupling?

3 7

parallel data computing in the alps code
by hamid mosadegh 13 Jul '17

13 Jul '17

Dear all For the MPS computing with a large bond dimension, I need to use parallel computing. But when I use the following commends, instead of a parallel data run, calculations are repeated several times. can alps do these calculations? Thanks for you attentions #setup environment export TOPDIR=$HOME/work/FINAL/ export MPIPATH=/share/binary/mpich2-1.4.1p1/bin/ export TMP=/home/scratch/ # generate nodelist cd $TOPDIR # setup run NP=8 #NAMDCONF=apoa1.namd # run $MPIPATH/mpirun -np $NP mps_optim --mpi --write-xml K.in.xml -- H. Mosadeq Shahr-e-Kord University (SKU), Shahr-e-Kord , Iran

2 1

Problem with DMRG and an Ising model
by Alexander Henry 12 Jul '17

12 Jul '17

Dear ALPS, I'm having some troubling getting a sensible result using DMRG with ALPS 2.3.0 on a simple transverse Ising model (i.e. using the "spin" model with Jz != 0, Gamma != 0). Other methods seem to get the correct result with the same parameters. It seems to be the Gamma term which causes the trouble. I've pasted a script below which demonstrates the problem. The script gives me the following output, with the second number being the ground state energy found by each method: Model fulldiag: {10.0: -5.3178022046739857} Model sparsediag: {10.0: -5.3178022046739937} Model mps_optim: {10.0: -5.3178022046739919} Model dmrg: {10.0: -213.49032836522568} Adjusting the various DMRG parameters produces similar results of E0 ~= -210. I assume I am doing something wrong, or perhaps the Gamma term isn't supported by DMRG? I have searched the mailing list archive without success. Ultimately, I want to use DMRG on a more complex model which includes a term similar to the Gamma / Sx term in the spin model. Thank you for your great work, Alex Henry The script: ########################## import pyalps import numpy as np import matplotlib.pyplot as plt import copy import math models = ['fulldiag', 'sparsediag', 'mps_optim', 'dmrg'] parms = [{ 'LATTICE' : "chain lattice", 'MODEL' : "spin", 'local_S' : 0.5, 'Jxy' : 0, 'Jz' : 1, 'Gamma' : 1, 'h' : 0, 'NUMBER_EIGENVALUES' : 1, 'L' : 10, }] def process(prefix, prog): input_file = pyalps.writeInputFiles(prefix, parms) res = pyalps.runApplication(prog, input_file) data = pyalps.loadEigenstateMeasurements(pyalps.getResultFiles(prefix=prefix)) return data def get_results(d): E0 = {} for Lsets in d: L = pyalps.flatten(Lsets)[0].props['L'] allE = [] for q in pyalps.flatten(Lsets): allE += list(q.y) allE = np.sort(allE) E0[L] = allE[0] return E0 results = {} for model in models: name = "ising_" + model data = process(name, model) results[model] = (get_results(data)) for model in models: print "Model " + model + ": " + str(results[model]) #############################

1 0

Long Range Kitaev Model site dependent Hamiltonian
by John Ian Kenneth E. Felismino 05 Jul '17

05 Jul '17

Hi! The hamiltonian I'm using is attached as a picture file. As you can see, the bond terms are proportional to abs(i - j)^\alpha. i was wondering if this is possible in Alps because I have tried doing it (crudely) with DMRG and found that it did not produce an output xml. I had simply added the term "/abs(i-j)^alpha#" in my hamiltonian. When I remove this term, the Hamiltonian works, so clearly the problem is with this proportionality term. Thanks. Ian Felismino

2 6

Multithreading in MacOSX
by H 03 Jul '17

03 Jul '17

Hello, I have installed alps via conda on a Linux OS (on a cluster) and on a Mac OS (on my desktop). I noticed that with the same code (using the MPS routines), on LinuxOS all the cores are utilized but on the MacOS only one core is used. Is there a method to make multithreading works on MacOS?

2 1