Re: [ALPS-users] dwa - bug? - Comp-phys-alps-users

17 Sep 2015


      Dear Tama,
I paste the files below.
If I run dwa like that
dwa conf.in.xml —Tmin=2 —Tmax=2
then the python script prints a 3x3 matrix for local density and local density^2
and if like this:
mpirun -np 10 dwa —mpi conf.in.xml —Tmin=2 —Tmax=2
then it prints a 9x1 vector
******* conf.in.xml *******
<?xml version="1.0" encoding="UTF-8"?>
<?xml-stylesheet type="text/xsl" href="ALPS.xsl"?>
<JOB xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:noNamespaceSchemaLocation="http://xml.comp-phys.org/2003/8/job.xsd">
  <OUTPUT file="conf.out.xml"/>
  <TASK status="new">
    <INPUT file="conf.task1.in.xml"/>
    <OUTPUT file="conf.task1.out.xml"/>
  </TASK>
</JOB>
******* conf.task1.in.xml *******
<?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="SKIP">32</PARAMETER>
<PARAMETER name="L">3</PARAMETER>
<PARAMETER name="MEASURE[Local Density]">True</PARAMETER>
<PARAMETER name="LATTICE">square lattice</PARAMETER>
<PARAMETER name="SEED">1504313687</PARAMETER>
<PARAMETER name="U">1.0</PARAMETER>
<PARAMETER name="T">0.04</PARAMETER>
<PARAMETER name="MODEL">boson Hubbard</PARAMETER>
<PARAMETER name="Nmax">5</PARAMETER>
<PARAMETER name="THERMALIZATION">5000</PARAMETER>
<PARAMETER name="SWEEPS">50000</PARAMETER>
<PARAMETER name="NONLOCAL">0</PARAMETER>
<PARAMETER name="mu">0.3828125</PARAMETER>
<PARAMETER name="MEASURE[Local Density^2]">True</PARAMETER>
<PARAMETER name="t">0.0595</PARAMETER>
  </PARAMETERS>
</SIMULATION>
********** python script *******
import pyalps
data = pyalps.loadMeasurements(pyalps.getResultFiles(prefix='conf’))
print data
...
On 16 Sep 2015, at 04:29, P. N. MA - Yotcopi Technologies tamama@yotcopi.com wrote:
Hi Matthias,
By default, I have disabled error bars for local densities. This is because for a more general case like 3D, the binned analysis use too much resource.
@Mateusz: can you send me your python script, for me to reproduce this error?
Thank you very much.
Best regards,
Tama Ma
Sent from my iPhone
...
On 16 Sep, 2015, at 3:20 am, Matthias Troyer troyer@phys.ethz.ch wrote:
Tama, can you please check why there are no error bars and it is a 2D array when running with one process and a 1D array with error bars when using more?
Matthias
...
On 15 Sep 2015, at 19:57, Mateusz Łącki mateusz.lacki@gmail.com wrote:
I would like to report something I thing is a bug.
If i measure “MEASURE[Local Density]”  for the “square lattice”, the result is (output of the  “.y” field of the :
[[ 0.99406836  1.00126848  0.99186967  ]
[ 0.99634558  0.98947165  0.99828454  ]
[ 0.98602864  0.99223814  0.98709175  ]]
and
1.10723308538 +/- 0.000732833211551
1.09709344128 +/- 0.000725375608498
1.10916580731 +/- 0.000737136950109
1.09653825096 +/- 0.000711748290787
1.09386893991 +/- 0.000713199643571
1.09495623341 +/- 0.000711331680729
1.09890852882 +/- 0.000721818241453
1.09269699362 +/- 0.000716545982708
1.08871501476 +/- 0.000706327757539
if I compute it using MPI.
I load the data using:
data = pyalps.loadMeasurements(pyalps.getResultFiles(prefix=pfix))
and I am speaking about proper element of the list data[0]  that of one for which s.props['observable']=='Local Density'
Please neglect different results as they come from different parameters.
Best,
Mateusz Łącki