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

Winding numbers
by "Michał Maik" 08 Feb '13

08 Feb '13

Hello, I asked this question earilier but recieved no response so I'll try again. I am curious if is it possible to get the winding number histogram from the SSE and worms code. There seem to be references to it in the code but since I am not familiar with C++ I do not know how to extract it. In Wmeas.C I found "measurements["Winding number histogram"]" but adding, "MEASURE[Winding number histogram]=true" to the parameter file doesn't do anything. If anyone has any suggestions on how to get this to work I would really apprieciate it. Thanks! Best Regards Michal Maik

2 1

to capture the reduced dm eigenvalues in dmrg
by A. Giovanni Ramirez Garcia 04 Feb '13

04 Feb '13

Hi all, Emanuele, I use to save the eigenvalues of the reduced density matrix too. The way I do it is to pipe the standard output to a grep command, then to a file. I'm using a GNU/Linux and there the grep command can "include" or "exclude" patterns it reads from a file. You know, the standard output for VERBOSE=1 includes too much information then I select what patterns I want to keep: SWEEP, ITERATION, RHO and I save that patterns in a file "patterns_file" (one per line) in the same directory where I put the python file with the definitions then, I run alpspython defs.py | grep -f patterns > rdm.log Then, the file rdm.log keeps the information about what SWEEP and what ITERATION has that list of RHO EIGENVALUES. Maybe it is not the best way to do it, I would like to know if someone else find another way to do it. Ciao Giovanni

1 0

Running MaxEnt code
by Viveka Singh 03 Feb '13

03 Feb '13

Dear all, I need to do the analytic continuation of my imaginary frequency / imaginary time data to the real frequency. I want to use the ALPS MaxEnt code for that, but I am not able to find an elaborate documentation on how to use the maxent code. I have looked at the "hybdoc.pdf", but not enough is being mentioned there and it is pointed there that a tutorial for the ALPS maxent code will be provided elsewhere. Can someone send me that link? Or can someone describe the input parameters that are required for the maxent code like - what are ALPHA's and how do one decide upon FREQUENCY_GRID? Do you need to spacify the OMEGA_MAX/OMEGA_MIN, even if someone wants to do the analytic continuation of imaginary time data? With regards, Viveka Nand Singh

2 1

Getting edge or bond type in spinmc
by Kourosh Sartipi 01 Feb '13

01 Feb '13

Hi everyone! I'm trying to write a code to do different calculations for different bond types, so I changed my lattice file and set edge type to 8888 for test, a part of lattice file is as follows: <EDGE type="8888"><SOURCE vertex="1"/><TARGET vertex="2" offset=" 0 0 0"/></EDGE> <EDGE type="8888"><SOURCE vertex="3"/><TARGET vertex="4" offset=" 0 0 0"/></EDGE> <EDGE type="8888"><SOURCE vertex="2"/><TARGET vertex="1" offset=" 1 0 0"/></EDGE> and I added a test code in alps/applications/mc/spins/spinsim.h as following: typedef typename AbstractSpinSim<MIdMatrix<double, M::dim> >::bond_iterator bond_iterator; for(bond_iterator bi = this->bonds().first; bi != this->bonds().second; bi++) { int et = this->edge_type(*bi); std::cout << et << std::endl; } but all I get from bond types is zero. Do you know how can I get the correct bond type here? Yours, Kourosh

1 0

reduced density matrix eigenvalues in DMRG
by Emanuele Levi 31 Jan '13

31 Jan '13

Dear ALPS users and developers, I am using the ALPS' DMRG application to run some simulations on a Heisenberg chain. Is there any way to output in the hdf file, or an external file, a certain number of eigenvalues of the reduced density matrix at each step? I know setting VERBOSE to 1 it prints in shell all the density matrix eigenvalues, and I could script this output. But is extremely long and not very convenient, so I am wandering if there is a better way of doing it. I thank you in advance for any help. Kindest regards. -- Emanuele Levi emanuele.levi(a)gmail.com

1 0

dirloop_sse density vs green's function measurement
by Francisco Cordobés 30 Jan '13

30 Jan '13

Dear all, I have been doing some tests for the Bose Hubbard model in 1D with dirloop_sse and I find that the the output density and the value of the Green's function at (0,0) do not coincide, as I would expect. I understand that when measuring Green's functions for the Bose-Hubbard model the output is that of <b^\dagger_0 b_j> with j going over all lattice points. b^\dagger_i is a creation operator and b_j is a destruction at sites i and j respectively. If that is correct, then, the following should be correct too and I should get <b^\dagger_0 b_0>=<n_0>=\rho. Where I denote the density by \rho. When I do the calculation, using the following input: <?xml version="1.0" encoding="UTF-8"?> <?xml-stylesheet type="text/xsl" href="ALPS.xsl"?> <SIMULATION xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:noNamespaceSchemaLocation="http://xml.comp-phys.org/2003/8/QMCXML.xsd"> <PARAMETERS> <PARAMETER name="MEASURE[Green Function]">True</PARAMETER> <PARAMETER name="MEASURE[Correlations]">True</PARAMETER> <PARAMETER name="L">64</PARAMETER> <PARAMETER name="LATTICE">chain lattice</PARAMETER> <PARAMETER name="SEED">189185290</PARAMETER> <PARAMETER name="U">10.0</PARAMETER> <PARAMETER name="T">0.5</PARAMETER> <PARAMETER name="MODEL">boson Hubbard</PARAMETER> <PARAMETER name="Nmax">3</PARAMETER> <PARAMETER name="THERMALIZATION">1000</PARAMETER> <PARAMETER name="SWEEPS">50000</PARAMETER> <PARAMETER name="NONLOCAL">0</PARAMETER> <PARAMETER name="mu">0.4</PARAMETER> <PARAMETER name="t">0.5</PARAMETER> </PARAMETERS> </SIMULATION> The output I get is: <SCALAR_AVERAGE name="Density"> <COUNT>55257</COUNT> <MEAN method="simple">0.6790808</MEAN> <ERROR converged="yes" method="binning">0.000306</ERROR> <VARIANCE method="simple">0.00258</VARIANCE> <AUTOCORR method="binning">0.505</AUTOCORR> <VECTOR_AVERAGE name="Green's Function" nvalues="64"> <SCALAR_AVERAGE indexvalue="( 0 ) -- ( 0 )"> <COUNT>55257</COUNT> <MEAN method="simple">1.178037</MEAN> <ERROR converged="yes" method="binning">0.000505</ERROR> <VARIANCE method="simple">0.0117</VARIANCE> <AUTOCORR method="binning">0.103</AUTOCORR> Which is very different. Am I interpreting the output correctly or am I just not running it long enough? Also, what is the meaning of the last three fields in the xml output field? n,n^2 and n^4? Because my initial guess would be to interpret n as the number of particles but it is not consistent with the density and the lattice size. Thank you very much, Francisco Cordobes .

2 2

hey
by Majid Abedi 29 Jan '13

29 Jan '13

have a look at this http://bit.ly/VQiD6S Majid

1 0

alps build error?
by Mateusz Łącki 29 Jan '13

29 Jan '13

Dear All, I am compiling alps and I am stuct at 64%. The error message is at the very bottom. My cmake line is : cmake28 -DCMAKE_INSTALL_PREFIX=~/opt/alps -D Boost_ROOT_DIR:PATH=~/opt/alps-2.1.1-r6176-src-with-boost/boost/ -DCMAKE_CXX_COMPILER=/opt/intel/composer_xe_2011_sp1.7.256/bin/intel64/icpc -DCMAKE_C_COMPILER=/opt/intel/composer_xe_2011_sp1.7.256/bin/intel64/icc -D LAPACK_64_BIT=ON -D HDF5_LIBRARIES=/opt/hdf5-1.8.8-intel/lib/libhdf5.so -D HDF5_INCLUDE_DIR=/opt/hdf5-1.8.8-intel/include/ -DALPS_BUILD_FORTRAN=ON ~/opt/alps-2.1.1-r6176-src-with-boost/alps/ compliler lines: -- The C compiler identification is Intel 12.1.0.20111011 -- The CXX compiler identification is Intel 12.1.0.20111011 -- The Fortran compiler identification is Intel What could be wrong ? I am using version alps-2.1.1-r6176-src-with-boost (straing from the web page, with FindLapack taken from the alps-2.1.1-r6670-src-with-boost). The version alps-2.1.1-r6670-src-with-boost fails at exactly the same point with exactly the same error message. Regards, Mateusz Lacki [ 64%] Building CXX object example/ietl/CMakeFiles/jacobidavidson2.dir/jacobidavidson2.cpp.o /home/lkb/lacki/opt/alps-2.1.1-r6176-src-with-boost/alps/src/boost/numeric/bindings/lapack/computational/getrf.hpp(197): error: no instance of overloaded function "boost::numeric::bindings::lapack::detail::getrf" matches the argument list argument types are: (order, ptrdiff_t, ptrdiff_t, double *, ptrdiff_t, int *) return detail::getrf( order(), bindings::size_row(a), ^ detected during: instantiation of "ptrdiff_t={long} boost::numeric::bindings::lapack::getrf_impl<Value>::invoke(MatrixA &, VectorIPIV &) [with Value=double, MatrixA=boost::numeric::ublas::matrix<double, boost::numeric::ublas::column_major, boost::numeric::ublas::unbounded_array<double, std::allocator<double>>>, VectorIPIV=std::vector<int, std::allocator<int>>]" at line 220 instantiation of "ptrdiff_t={long} boost::numeric::bindings::lapack::getrf(MatrixA &, VectorIPIV &) [with MatrixA=boost::numeric::ublas::matrix<double, boost::numeric::ublas::column_major, boost::numeric::ublas::unbounded_array<double, std::allocator<double>>>, VectorIPIV=std::vector<int, std::allocator<int>>]" at line 285 of "/home/lkb/lacki/opt/alps-2.1.1-r6176-src-with-boost/alps/src/ietl/jd.h" instantiation of "void ietl::solver::left_prec_solver<SOLV, MATRIX, VS, PREC>::operator()(ietl::real_type<ietl::vectorspace_traits<VS>::scalar_type>::type, ietl::vectorspace_traits<VS>::vector_type &, ietl::vectorspace_traits<VS>::vector_type &, IT &) [with SOLV=ietl::gmres_wrapper, MATRIX=matrix_t, VS=ietl::vectorspace<vector_t>, PREC=jacobi_prec<matrix_t, double>, IT=ietl::jd_iteration<double>]" at line 611 of "/home/lkb/lacki/opt/alps-2.1.1-r6176-src-with-boost/alps/src/ietl/jd.h" instantiation of "void ietl::jd<MATRIX, VS>::jdqr(SOLVER &, IT &, GEN &, ietl::vectorspace_traits<VS>::size_type, bool) [with MATRIX=matrix_t, VS=ietl::vectorspace<vector_t>, SOLVER=ietl::solver::left_prec_solver<ietl::gmres_wrapper, matrix_t, ietl::vectorspace<vector_t>, jacobi_prec<matrix_t, double>>, IT=ietl::jd_iteration<double>, GEN=gen_t]" at line 384 of "/home/lkb/lacki/opt/alps-2.1.1-r6176-src-with-boost/alps/src/ietl/jd.h" instantiation of "void ietl::jd<MATRIX, VS>::eigensystem(IT &, GEN &, ietl::vectorspace_traits<VS>::size_type, PREC &, SOLV, bool) [with MATRIX=matrix_t, VS=ietl::vectorspace<vector_t>, SOLV=ietl::gmres_wrapper, IT=ietl::jd_iteration<double>, GEN=gen_t, PREC=jacobi_prec<matrix_t, double>]" at line 165 of "/home/lkb/lacki/opt/alps-2.1.1-r6176-src-with-boost/alps/example/ietl/jacobidavidson2.cpp" /home/lkb/lacki/opt/alps-2.1.1-r6176-src-with-boost/alps/src/boost/numeric/bindings/lapack/computational/getrs.hpp(222): error: no instance of overloaded function "boost::numeric::bindings::lapack::detail::getrs" matches the argument list argument types are: (order, trans, ptrdiff_t, ptrdiff_t, const double *, ptrdiff_t, const int *, double *, ptrdiff_t) return detail::getrs( order(), trans(), bindings::size_column_op(a, ^ detected during: instantiation of "ptrdiff_t={long} boost::numeric::bindings::lapack::getrs_impl<Value>::invoke(const MatrixA &, const VectorIPIV &, MatrixB &) [with Value=double, MatrixA=boost::numeric::ublas::matrix<double, boost::numeric::ublas::column_major, boost::numeric::ublas::unbounded_array<double, std::allocator<double>>>, VectorIPIV=std::vector<int, std::allocator<int>>, MatrixB=boost::numeric::ublas::vector<double, boost::numeric::ublas::unbounded_array<double, std::allocator<double>>>]" at line 247 instantiation of "ptrdiff_t={long} boost::numeric::bindings::lapack::getrs(const MatrixA &, const VectorIPIV &, MatrixB &) [with MatrixA=boost::numeric::ublas::matrix<double, boost::numeric::ublas::column_major, boost::numeric::ublas::unbounded_array<double, std::allocator<double>>>, VectorIPIV=std::vector<int, std::allocator<int>>, MatrixB=boost::numeric::ublas::vector<double, boost::numeric::ublas::unbounded_array<double, std::allocator<double>>>]" at line 294 of "/home/lkb/lacki/opt/alps-2.1.1-r6176-src-with-boost/alps/src/ietl/jd.h" instantiation of "void ietl::solver::left_prec_solver<SOLV, MATRIX, VS, PREC>::operator()(ietl::real_type<ietl::vectorspace_traits<VS>::scalar_type>::type, ietl::vectorspace_traits<VS>::vector_type &, ietl::vectorspace_traits<VS>::vector_type &, IT &) [with SOLV=ietl::gmres_wrapper, MATRIX=matrix_t, VS=ietl::vectorspace<vector_t>, PREC=jacobi_prec<matrix_t, double>, IT=ietl::jd_iteration<double>]" at line 611 of "/home/lkb/lacki/opt/alps-2.1.1-r6176-src-with-boost/alps/src/ietl/jd.h" instantiation of "void ietl::jd<MATRIX, VS>::jdqr(SOLVER &, IT &, GEN &, ietl::vectorspace_traits<VS>::size_type, bool) [with MATRIX=matrix_t, VS=ietl::vectorspace<vector_t>, SOLVER=ietl::solver::left_prec_solver<ietl::gmres_wrapper, matrix_t, ietl::vectorspace<vector_t>, jacobi_prec<matrix_t, double>>, IT=ietl::jd_iteration<double>, GEN=gen_t]" at line 384 of "/home/lkb/lacki/opt/alps-2.1.1-r6176-src-with-boost/alps/src/ietl/jd.h" instantiation of "void ietl::jd<MATRIX, VS>::eigensystem(IT &, GEN &, ietl::vectorspace_traits<VS>::size_type, PREC &, SOLV, bool) [with MATRIX=matrix_t, VS=ietl::vectorspace<vector_t>, SOLV=ietl::gmres_wrapper, IT=ietl::jd_iteration<double>, GEN=gen_t, PREC=jacobi_prec<matrix_t, double>]" at line 165 of "/home/lkb/lacki/opt/alps-2.1.1-r6176-src-with-boost/alps/example/ietl/jacobidavidson2.cpp" compilation aborted for /home/lkb/lacki/opt/alps-2.1.1-r6176-src-with-boost/alps/example/ietl/jacobidavidson2.cpp (code 2) make[2]: *** [example/ietl/CMakeFiles/jacobidavidson2.dir/jacobidavidson2.cpp.o] Erreur 2 make[1]: *** [example/ietl/CMakeFiles/jacobidavidson2.dir/all] Erreur 2 make: *** [all] Erreur 2

3 3

FindLapack.cmake
by Mateusz Łącki 25 Jan '13

25 Jan '13

Dear All I have found the following code in FindLapack.cmake trying to do installation of new alps: IF($ENV{MKL} MATCHES "mkl") MESSAGE(STATUS "Using intel/mkl library: $ENV{MKL}") #look for the path where the mkl is installed 43: STRING(REGEX MATCHALL "[-][L]([^ ;])+" MKL_PATH_WITH_PREFIX "$ENV{MKL}") STRING(REGEX REPLACE "[-][L]" "" MKL_PATHS ${MKL_PATH_WITH_PREFIX}) ENDIF($ENV{MKL} MATCHES "mkl") If I assume correctly this code assumes that there is an environment variable dalled MKL which contains something like "-L/path/to/mkl" on my platform it is set by default to "/path/to/mkl" If I understand it correctly at the line 43, the $MKL gets stripped of "-L" part and the ${MKL_PATH_WITH_PREFIX} then will contain just "/path/to/mkl". I think that perhaps some sort of conditional clause could be added so that both $MKL = /path/to/mkl and $MKL = -L/path/to/mkl worked. Regards, Mateusz Lacki

2 3

alps installation: lapack vs intel/mkl
by Mateusz Łącki 25 Jan '13

25 Jan '13

Dear All, I would like to ask about the lapack library used by alps: I have a following output: -- Using intel/mkl library: -L/opt/intel/composer_xe_2011_sp1.7.256/mkl MKL_PATH_WITH_PREFIX -L/opt/intel/composer_xe_2011_sp1.7.256/mkl -- Looking for intel/mkl library in /opt/intel/composer_xe_2011_sp1/mkl//lib/intel64;/opt/intel/composer_xe_2011_sp1/mkl//lib/emt;/opt/intel/composer_xe_2011_sp1/mkl//lib/;/opt/intel/composer_xe_2011_sp1/mkl//lib;/opt/intel/composer_xe_2011_sp1.7.256/mkl Falling back to CMake provied LAPACK/BLAS detection. -- A library with BLAS API found. -- A library with BLAS API found. -- A library with LAPACK API not found. Please specify library location. (i have uncommented debug messages) so my mkl is located in: /opt/intel/composer_xe_2011_sp1/mkl//lib/intel64 which is btw the default localtion of mkl library in newer versions of the compiler in the FindLapack.cmake the lines 61-65 are: STRING(REGEX MATCH "[0-9]+\\.[0-9]+\\.[0-9]+" MKL_VERSION "$ENV{MKL_HOME}") SET(LINK_LAPACK_DEFAULT 1) IF(${MKL_VERSION} MATCHES "10\\.0\\.[0-2]") SET(LINK_LAPACK_DEFAULT 0) ENDIF(${MKL_VERSION} MATCHES "10\\.0\\.[0-2]") I think (but I have never ever used cmake in my project) that this lines assume that mkl s installed in a directory that that contains specific version: /opt/intel/Compiler/11.5.6/mkl (then MKL_Version becomes 11.5.6) which is often not the case. I will try to fix this cmake file as well as I can... Regards, Mateusz Lacki

1 0