Comp-phys-alps-users March 2015

comp-phys-alps-users@lists.phys.ethz.ch

6 participants
6 discussions

Inconsistent comment in generate_spline_matrix in the DMFT folder
by Kuangshing Chen 22 Jul '15

22 Jul '15

Hi, The function generate_spline_matrix in fouriertransform.C in the dmft folder shows the comments as follows: // A is the matrix whose inverse defines spline_matrix // // 6 6 // 1 4 1 // 1 4 1 // A = ... // // 1 4 1 // -2 0 0 2 However, the following codes, dense_matrix A = 4*dt/6.*boost::numeric::ublas::identity_matrix<double>(Np1); for (int i=1; i<Np1-1; i++) { A(i,i-1) = dt/6.; A(i,i+1) = dt/6.; } A(0,0) = 1.; A(0, Np1-1) = 1.; A(Np1-1, 0) = -2.*dt/6.; A(Np1-1, 1) = -1.*dt/6.; A(Np1-1, Np1-2) = 1*dt/6.; A(Np1-1, Np1-1) = 2*dt/6.; the red lines will make the matrix looks like // 6 6 // 1 4 1 // 1 4 1 // A = ... // // 1 4 1 // -2 *-1* *1* 2 Is the comment correct or the code correct? Any reference for this algorithm? Thank you, Kuang-Shing Chen

5 25

sparsediag on lattice with different edge types
by Valentin Sinitsyn 24 Mar '15

24 Mar '15

Hi all, I'm trying to do exact diagonalization (with sparsediag) for a spin tube (see lattice description below). Everything goes fine, but energy I get is about twice higher (by absolute value) than expected (I can treat small lattices with other means). This looks suspiciously similar to [1], but that message was left unanswered for two years. Am I doing anything wrong or is this a bug in sparsediag? For the latter, I can help with debugging. Thank you in advance! 1. https://lists.phys.ethz.ch/pipermail/comp-phys-alps-users/2013/002100.html Params file: LATTICE_LIBRARY="./my_lattice.xml" LATTICE="ladder" W=4 MODEL="spin"; CONSERVED_QUANTUMNUMBERS="Sz"; TRANSLAION_SYMMETRY=false; local_S=1/2; J1=1.0; J2=-0.0625; J3=-0.125; Sz_total=0 {L=6} Sz_total=1 {L=6} Lattice description is as follows: <LATTICES> <LATTICE name="square lattice" dimension="2"> <PARAMETER name="a" default="1"/> <BASIS><VECTOR>a 0</VECTOR><VECTOR>0 a</VECTOR></BASIS> </LATTICE> <FINITELATTICE name="rectangular half-open" dimension="2"> <LATTICE ref="square lattice"/> <EXTENT dimension="1" size="L"/> <EXTENT dimension="2" size="W"/> <BOUNDARY dimension="1" type="open"/> <BOUNDARY dimension="2" type="periodic"/> </FINITELATTICE> <UNITCELL name="nnn" dimension="2"> <VERTEX/> <EDGE type="1"> <SOURCE vertex="1"/> <TARGET vertex="1" offset="1 0"/> </EDGE> <EDGE type="2"> <SOURCE vertex="1"/> <TARGET vertex="1" offset="0 2"/> </EDGE> <EDGE type="3"> <SOURCE vertex="1"/> <TARGET vertex="1" offset="0 1"/> </EDGE> </UNITCELL> <LATTICEGRAPH name="ladder"> <FINITELATTICE ref="rectangular half-open"/> <UNITCELL ref="nnn"/> </LATTICEGRAPH> </LATTICES> -- Regards, Valentine Sinitsyn

2 3

MPS with odd number of sites
by Amir M.Aghaei 23 Mar '15

23 Mar '15

Hi, Is it possible to use odd number of sites with the MPS code? I am getting the following error when I change the number of sites from 8 to 7 using the tutorial file (with almost no change but changing the lattice and lattice libraries, lattice and python files are attached in case you want to reproduce): Running task 1. > This binary contains symmetries: nu1 > MAQUIS repo r4143 > Temporary storage is disabled > site_types: 0 0 0 0 0 0 0 > phys[0]: |( <1,0,0,0,0,0>: 1 )( <-1,0,0,0,0,0>: 1 )| > Right end: <0,0,0,0,0,0> > MPS initialization has finished... > Expecting 0 states to orthogonalize to. > Boundaries are partially initialized... > Boundaries are fully initialized... > Done init_left_right > TSMPOTensor: 1x5, 5 operators, 8 tags > TSMPOTensor: 5x5, 8 operators, 11 tags > TSMPOTensor: 5x5, 9 operators, 11 tags > TSMPOTensor: 5x5, 9 operators, 11 tags > TSMPOTensor: 5x5, 9 operators, 11 tags > TSMPOTensor: 5x1, 5 operators, 8 tags > Total number of tags: 60 > > Sweep 0, optimizing sites 0 and 1 > Exception thrown: > vector::_M_fill_insert > The group '/spectrum' does not exist. > regards, Amir

2 3

Building ALPS from source
by Tadeusz Wasiutyński 19 Mar '15

19 Mar '15

Some time ago I have installed alps-2.2.b3-r7462 on desktop running Fedora 21. Everything went smoothly. Now I wanted to do the same on the other comp. Again seems that cmake went properly, however make collapses at 19%. What is the reason? I assume that the alps2.2.b3-r7462 is the same. regards Tadeusz Wasiutyński

1 0

Measuring correlations only between some, given, sites
by Amir M.Aghaei 09 Mar '15

09 Mar '15

Hi, I have gone through the tutorial and the mailing list archive but I was unable to find whether it is possible to measure the correlation only between some sites of the lattice under study, say about O(n) of possible correaltions not every possible 2 sites in the lattice which is about O(n^2) and is definitely a waste of calculations in my case. So, is there any way that I can give a list (of two sites) to the program? If not, can you think of a very easy workaround? My main interest right now is the new DMRG MPS code, so even if there is any MPS specific technique that I can use I would be happy to hear about it. Bests, Amir

2 3

hdf5 outpute
by bondnina 04 Mar '15

04 Mar '15

Dear Matthias and others! Again need your help. So finally I have performed installation on our cluster. Everything works fine!!! I have run tutorial mps-05-bhquench. I have got lots of results h5 files. But watching them in the viewer is not so user-friendly. Is there any tutorial how to decrypt the output. Best, Nina -- Yours sincerely, Nina Bondarenko 1024R/EAC56154 9F6C DDF8 6873 5D19 FF9A 600E 46A4 42BE EAC5 6154

1 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Comp-phys-alps-users March 2015