Now that we have a public release of IMP that non-developers can use,
there is a new imp-users mailing list. This is targeted at "users" of
IMP, i.e. for discussion of and help with the use of IMP, as opposed to
technical discussion of new features.
Note that I did not automatically subscribe all the imp-dev people to
imp-users. However, I suggest that you do subscribe to the imp-users
list yourselves: http://salilab.org/imp/nightly/doc/html/maillists.html
The non-technical amongst you might also want to unsubscribe from
imp-dev at the same time!
Ben
--
ben(a)salilab.org http://salilab.org/~ben/
"It is a capital mistake to theorize before one has data."
- Sir Arthur Conan Doyle
https://salilab.org/imp/bugs/show_bug.cgi?id=106
--- Comment #4 from drussel(a)SALILAB.ORG 2010-04-20 13:02:33 PDT ---
(In reply to comment #3)
> If by "test bed" you mean a script which demonstrates the problem, the attached
> memtest.py does exactly that.
Yeah, I meant "now that we have" :-)
>
> If you want to explore other ways of solving the problem, you can chat with
> Harianto Tjong over on imp-users, who I'm sure will be willing to help try
> things out.
--
Configure bugmail: https://salilab.org/imp/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
https://salilab.org/imp/bugs/show_bug.cgi?id=106
--- Comment #3 from Ben Webb <ben(a)salilab.org> 2010-04-20 12:46:47 PDT ---
If by "test bed" you mean a script which demonstrates the problem, the attached
memtest.py does exactly that.
If you want to explore other ways of solving the problem, you can chat with
Harianto Tjong over on imp-users, who I'm sure will be willing to help try
things out.
--
Configure bugmail: https://salilab.org/imp/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
https://salilab.org/imp/bugs/show_bug.cgi?id=106
--- Comment #2 from drussel(a)SALILAB.ORG 2010-04-20 12:43:28 PDT ---
I see two easy ways forward:
1. convert the bad_alloc into a python-friendly exception with a nicer message.
I'm not sure how useful this is as I suspect a general purpose optimizer and
100k particles and a nonbonded list can't be stably mixed with our current
setup as it is likely to hit a tight configuration at some point.
2. convert the bad_alloc to a ModelException. Then if MC is being run, it can
reject the conformation or CG run and try another so at least something can
come out of it. But it makes the assumption that such a configuration would
have a bad score.
Slightly more involved, which I used to have in there is to reduce the slack in
the NBL and try again. If we have a test bed, then I can experiment with that.
--
Configure bugmail: https://salilab.org/imp/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
dear all,
i have been trying to run the example for the non-bonded list in IMP
(the official release 1.0) and sadly, it did not work:
sendling 1026% /fs/pool/pool-foerster/apps/src/imp/branches/1.0/tools/imppy.sh
python test_imp.py
Traceback (most recent call last):
File "test_imp.py", line 10, in <module>
ps = IMP.core.ListSingletonContainer(IMP.core.create_xyzr_particles(m,
20, 1.0))
AttributeError: 'module' object has no attribute 'ListSingletonContainer'
test_imp.py is the IMP example:
import IMP
import IMP.core
import IMP.atom
# This example addes a restraint on nonbonded interactions
# after excluding a set of bonded interactions.
m= IMP.Model()
# The set of particles
ps = IMP.core.ListSingletonContainer(IMP.core.create_xyzr_particles(m, 20, 1.0))
# create a bond between two particles
bd0= IMP.atom.Bonded.setup_particle(ps.get_particle(0))
bd1= IMP.atom.Bonded.setup_particle(ps.get_particle(1))
IMP.atom.custom_bond(bd0, bd1, 2.0)
# Set up the nonbonded list for all pairs at are touching
# and let things move 3 before updating the list
nbl= IMP.core.ClosePairContainer(ps, 0.0, 3.0)
nbl.add_pair_filter(IMP.atom.BondedPairFilter())
# Set up excluded volume
sdps= IMP.core.SphereDistancePairScore(IMP.core.HarmonicLowerBound(0,1))
evr= IMP.core.PairsRestraint(sdps, nbl)
m.add_restraint(evr)
# Set up optimizer
o= IMP.core.ConjugateGradients()
o.set_model(m)
o.optimize(1000)
any suggestions?
thanks
p.s.: IMP is compiled with build='fast'.
all 'required' packages (specified in the installation doc) are installed:
* scons (0.98 or later)
* Boost (1.33 or later)
* A Subversion (svn) client to access the repository
If you wish to build the Python interfaces, you will also need:
* Python (2.3 or later)
* SWIG (1.3.40 or later)
--
Dr. Friedrich Foerster
Max-Planck Institut fuer Biochemie
Am Klopferspitz 18
D-82152 Martinsried
Tel: +49 89 8578 2632
Fax: +49 89 8578 2641
www.biochem.mpg.de/foerster
