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

Unable to run looper QMC simulation
by Michelle 18 May '20

18 May '20

2 1

Unequal-time Green function in SSE QMC
by 郭怀明 11 Apr '20

11 Apr '20

Dear all, It seems to me that the realization of SSE QMC in ALPS does not measure the unequal-time Green function. Is there any possibility to revise the source code to add such a meansurement? Sincerely, Huaiming Guo

2 2

eigenvectors
by Negin Mohseni 08 Apr '20

08 Apr '20

Dear All, Firstly I thank Rongyang for his fruitful explanation and suggestions. He assured me that I had thought correctly. In fact, in our work I must have all eigenvectors of a system. For this reason, I read tutorials related to exact diagonalization (ED) from sparse diagonalization to full diagonalization. Unfortunately, none of them fulfills my need. In most of tutorials the energy eigenvalues are sorted in several subspaces and momentums categories. But, I just wish to have a data set containing eigenvectors ignoring the dependence of momentum or subspaces. For example, for Ising chain in a transverse field ("h"), I wish to have the system's eigenvectors (as a list) and corresponded eigenvalues for each "h". How can we able to have a system's eigenvectors and corresponded eigenvalues? All the best

3 2

Error in QMC with worm agrithm
by 郭怀明 07 Apr '20

07 Apr '20

Dear all, While I try to perform simulation using QMC with worm agrithm, I met an error related to HDF5. The information is attached as a file. I wonder how to fix the problem. Thank you for the helps! Sincerely, Huaiming Guo

3 3

eigenvector from DMRG
by Negin Mohseni 06 Apr '20

06 Apr '20

I wish good health for all developers and users these days. My question is about DMRG process, We know that in DMRG procedure, while the size of the system grows, just the most important eigenvectors are kept (in each iteration). I think the 'MAXSTATES' in the DMRG code of ALPS package illustrates the maximum number of states from diagonalizing the reduced density matrix. Since the number of such states is less than the number of our system's state then we have an error in estimating energy of the system. But a question has occupied my mind: Does it make sense to ask the code to give the ground state eigenvector? Because with keeping a few numbers of the system's eigenvectors, I think in the last iteration, the output eigenvector is not corresponded to the whole system from the size point of view. We can consider a chain with 30 spin-1/2 particles. The ground state eigenvector of the system must be 2^30 in length . But I think the eigenvector of the last iteration from DMRG code is very smaller than 2^30 in length. Is this statement true? If it is correct, then how we can have the ground state eigenvector from ALPS DMRG?

2 1

extracting outputs
by Negin Mohseni 24 Mar '20

24 Mar '20

Dear All, My question is related to the calculated results of the below code. This entails MPS implantation to reach energy gap of a Heisenberg chain in the presence of a magnetic field. We wrote: %%%%%%%%%%%%%%% Start %%%%%%%%%%%%%%%%%%%%%%% import pyalps import numpy as np import matplotlib.pyplot as plt import pyalps.plot parms = [] for hz in [0.0, 0.2, 0.4, 0.6, 0.8, 1.0]: parms.append( { 'LATTICE' : "open chain lattice", 'MODEL' : "spin", 'CONSERVED_QUANTUMNUMBERS' : 'N,Sz', 'Sz_total' : 0, 'J' : 1, 'h' : hz, 'SWEEPS' : 4, 'L' : 32, 'MAXSTATES' : 100, 'NUMBER_EIGENVALUES' : 2 } ) input_file = pyalps.writeInputFiles('parm_spin_one_half_gap',parms) res = pyalps.runApplication('mps_optim',input_file,writexml=True) data = pyalps.loadEigenstateMeasurements(pyalps.getResultFiles(prefix='parm_spin_one_half_gap')) energies = np.empty(0) for s in data[0]: if s.props['observable'] == 'Energy': energies = s.y else: print(s.props['observable'], ':', s.y[0]) energies.sort() print('Energies:', end=' ') for e in energies: print(e, end=' ') print('\nGap:', abs(energies[1]-energies[0])) %%%%%%%%%%%%%%% END %%%%%%%%%%%%%%%%%%%%%%% Although, in the file "parm_spin_one_half_gap.in.xml" there are expensive information such as energies for any "hz", we need to collect all results in a *.dat or *.txt file. For instance, a "file.txt" as: hz gap 0.0 0.11 0.2 0.12 0.4 0.14 0.6 0.15 How can we generate such a file?!

1 0

ALPS installation via Conda and SVN Trunk package
by Manpreet Singh 10 Mar '20

10 Mar '20

Dear all, I have installed ALPS via Conda on Linux Ubuntu 16.04 LTS and it seems to work fine. However, there is an ALPS Trunk version available via SVN at alps.comp-phys.org/svn/alps1/trunk/alps. I would like to know which one of these versions has the latest and updated libraries. Also, if ALPS Trunk is the latest version, how do I update the existing libraries which were installed via Conda? Any help would be appreciated. Thanks in advance! Best regards- Manpreet

2 1

drawing an arbitrary lattice
by Negin Mohseni 01 Mar '20

01 Mar '20

So sorry there was a mistake in the last email. I has corrected recently and asked it again. Dear all Firstly, thanks to Matthias for his previous answers. We are looking for a way to map our lattice. With guidance presented in the web site we wrote our lattice. Although, the unit cell of what we wanted has been brought by default in the lattice collections of the ALPS, we are trying to apply and see what is in our mind explicitly. We wrote: <LATTICES> <LATTICE name="stripm" dimension="2"> <PARAMETER name="a" default="1.73205"/> <PARAMETER name="b" default="1.73205"/> <PARAMETER name="phi" default="Pi/3"/> <BASIS> <VECTOR> a 0 </VECTOR> <VECTOR> b*cos(phi) b*sin(phi) </VECTOR> </BASIS> </LATTICE> <FINITELATTICE name="honeyladder" dimension="2"> <LATTICE ref="stripm"/> <PARAMETER name="L" default="10"/> <PARAMETER name="W" default="2" /> <EXTENT dimension="1" size="L" /> <EXTENT dimension="2" size="W" /> <BOUNDARY dimension="1" type="periodic" /> <BOUNDARY dimension="2" type="open" /> </FINITELATTICE> <UNITCELL name="honeyl" dimension="2" vertices="2"> <VERTEX id="1" ><COORDINATE> cos(phi/2) cos(phi/2) </COORDINATE></VERTEX> <VERTEX id="2"><COORDINATE> 2*cos(phi/2) 2* cos(phi/2) </COORDINATE></VERTEX> <EDGE><SOURCE vertex="1"/><TARGET vertex="2"/></EDGE> <EDGE><SOURCE vertex="2"/><TARGET vertex="1" offset="1 0"/></EDGE> <EDGE><SOURCE vertex="2"/><TARGET vertex="1" offset="0 1"/></EDGE> </UNITCELL> <LATTICEGRAPH> <FINITELATTICE ref="honeyladder"/> <UNITCELL ref="honeyl"/> </LATTICEGRAPH> </LATTICES> *Question*: How to see the output of the .xml file? Thanks in advance

2 1

draw an arbitrary lattice
by Negin Mohseni 01 Mar '20

01 Mar '20

Dear all Firstly, thanks to Matthias for his previous answers. We are looking for a way to map our lattice. With guidance presented in the web site we wrote our lattice. Although, the unit cell of what we wanted has been brought by default in the lattice collections of the ALPS, we are trying to apply and see what is in our mind explicitly. We wrote: <LATTICES> <LATTICE name="stripm" dimension="2"> <PARAMETER name="a" default="1.73205"/> <PARAMETER name="b" default="1.73205"/> <PARAMETER name="phi" default="Pi/3"/> <BASIS> <VECTOR> a 0 </VECTOR> <VECTOR> b*sin(phi) b*cos(phi) </VECTOR> </BASIS> </LATTICE> <FINITELATTICE name="honeyladder" dimension="2"> <LATTICE ref="stripm"/> <PARAMETER name="L" default="10"/> <PARAMETER name="W" default="2" /> <EXTENT dimension="1" size="L" /> <EXTENT dimension="2" size="W" /> <BOUNDARY dimension="1" type="periodic" /> <BOUNDARY dimension="2" type="open" /> </FINITELATTICE> <UNITCELL name="honeyl" dimension="2" vertices="2"> <VERTEX id="1" ><COORDINATE> cos(phi) cos(phi) </COORDINATE></VERTEX> <VERTEX id="2"><COORDINATE> 2*cos(phi) 2* cos(phi) </COORDINATE></VERTEX> <EDGE><SOURCE vertex="1"/><TARGET vertex="2"/></EDGE> <EDGE><SOURCE vertex="2"/><TARGET vertex="1" offset="1 0"/></EDGE> <EDGE><SOURCE vertex="2"/><TARGET vertex="1" offset="0 1"/></EDGE> </UNITCELL> <LATTICEGRAPH> <FINITELATTICE ref="honeyladder"/> <UNITCELL ref="honeyl"/> </LATTICEGRAPH> </LATTICES> *Question*: How to see the output of the .xml file? Thanks in advance

1 0

Restarting simulation after unexpected shutdown
by Robertson Esperanza 01 Mar '20

01 Mar '20

Hi, I am wondering if there is a way to re-run the simulation of multiple tasks (I am using mps_optim) without repeating all the measurements all the way from task 1, especially when I already have a significant number of tasks finished before simulation unexpectedly halted (i.e. due to an unexpected shutdown). In my naive way, rerunning the simulation (when the checkpoint file is retained), measurements are repeated from task 1 until the task where the dmrg simulation stopped say at task (x) (and this will take time), and then dmrg sweeps for task (x+1) will proceed as usual. Thanks, Robertson C. Esperanza

2 1