Giter Club home page Giter Club logo

Comments (6)

 avatar commented on June 4, 2024

Here's the compressed version of the CB8.xyz input file that I used for the above run:
CB8.xyz.zip

If you're tempted to reproduce, please be aware that the run took 7.5 hrs with 12 threads on a 12-core mid2010 Mac Pro

from crest.

pprcht avatar pprcht commented on June 4, 2024

I don't think it is related to OMP stacksizes and -alpb. At this point where it crashes all calculations are done and only I/O is left. Have you checked the structures in crest_property.xyz? Maybe they broke and hence some reading/writing does not work

from crest.

 avatar commented on June 4, 2024

The example above is cucurbit[8]uril. There are only two structures in crest_property.xyz and they both seem to be intact - one is d8h symmetry, the other cs, at much higher energy, as expected. This same job (with -gbsa instead of -alpb) worked fine in 2.10.2.

In issue #15 you pointed out that the CREGEN data ends up in PROP/cregen.out.tmp. This certainly is the case for me with small guest molecules, but not with the larger CB8 host. Perhaps an array bound is being inadvertently violated during i/o now?

from crest.

pprcht avatar pprcht commented on June 4, 2024

As mentioned in the other issue I have updated the attached binary of the pre-release, but this problem/bug here still eludes me. However, the crest_property.xyz should be the final ensemble and all that is missing is sorting it, so maybe you could try to use crest coord --cregen crest_property.xyz. Although with only two structures the outcome of this is probably expectable...

from crest.

 avatar commented on June 4, 2024

Thanks for sharing the updated version but it still stops at the same place (this time with 3 structures). The final lines in the output file are:

 ---------------------------------------
 Hessian calculations for all conformers
 ---------------------------------------
 Performing calculations for 3 structures ...
 1 2 3 
 done.

The separate cregen job on PROP/crest_property.xyz runs OK so this gives me the thermo data:

 Command line input:
 > crest CB8.xyz --cregen PROP/crest_property.xyz

  --cregen : CREGEN standalone usage. Sorting file <PROP/crest_property.xyz>
 =============================
  # threads =          12
 =============================
 Using only the cregen sorting routine.
 input  file name : PROP/crest_property.xyz
 output file name : PROP/crest_property.xyz.sorted
 number of atoms                :   144
 number of points on xyz files  :   3
 RMSD threshold                 :   0.1250
 Bconst threshold               :   0.0100
 population threshold           :   0.0500
 conformer energy window  /kcal :   6.0000
 # fragment in coord            :     1
 # bonds in reference structure :   168
 number of reliable points      :     3
 reference state Etot :  -286.136982583200     
 running RMSDs...
 done.
 number of doubles removed by rot/RMSD         :           0
 total number unique points considered further :           3
       Erel/kcal        Etot weight/tot  conformer     set   degen     origin
       1   0.000  -286.13698    0.99352    0.99352       1       1     
       2   3.014  -286.13218    0.00617    0.00617       2       1     
       3   4.778  -286.12937    0.00031    0.00031       3       1     
T /K                                  :   298.15
E lowest                              :  -286.13698
ensemble average energy (kcal)        :    0.020
ensemble entropy (J/mol K, cal/mol K) :    0.336    0.080
ensemble free energy (kcal/mol)       :   -0.024
population of lowest in %             :   99.352
 number of unique conformers for further calc            3
 list of relative energies saved as "crest.energies"
 
 -----------------
 Wall Time Summary
 -----------------
              CREGEN wall time :         0h : 0m : 0s
--------------------
Overall wall time  : 0h : 0m : 0s
 
 CREST terminated normally.

I also ran the same -nci job with a larger 1+ guest@CB8 complex (2 components) and it stops at the same place after performing hessian calcs on 26 ensemble structures:

 writing TMPCONF* Dirs from file "crest_conformers.xyz" ... done.
 ---------------------------------------
 Hessian calculations for all conformers
 ---------------------------------------
 Performing calculations for 26 structures ...
 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 
 done.

HOWEVER this time the cregen job incorrectly calculates the thermo data. Note that only one (not two) fragments are detected, and all 26 structures are removed from the ensemble as 'clashes' :(

 Command line input:
 > crest [email protected] -chrg 1 --cregen PROP/crest_property.xyz

  -chrg 1
  --cregen : CREGEN standalone usage. Sorting file <PROP/crest_property.xyz>
 =============================
  # threads =          12
 =============================
 Using only the cregen sorting routine.
 input  file name : PROP/crest_property.xyz
 output file name : PROP/crest_property.xyz.sorted
 number of atoms                :   171
 number of points on xyz files  :   26
 RMSD threshold                 :   0.1250
 Bconst threshold               :   0.0100
 population threshold           :   0.0500
 conformer energy window  /kcal :   6.0000
 # fragment in coord            :     1
 number of removed clashes      :    26
 # bonds in reference structure :   198
 number of reliable points      :     0
 reference state Etot :  1.158413105592492E-310
 running RMSDs...
 done.
 number of doubles removed by rot/RMSD         :           0
 total number unique points considered further :           0
       Erel/kcal        Etot weight/tot  conformer     set   degen     origin
T /K                                  :   298.15
E lowest                              :     0.00000
ensemble average energy (kcal)        :    0.000
ensemble entropy (J/mol K, cal/mol K) :    0.000    0.000
ensemble free energy (kcal/mol)       :    0.000
population of lowest in %             :    0.000
 number of unique conformers for further calc            0
 list of relative energies saved as "crest.energies"
 
 -----------------
 Wall Time Summary
 -----------------
              CREGEN wall time :         0h : 0m : 0s
--------------------
Overall wall time  : 0h : 0m : 0s
 
 CREST terminated normally.

If you want to add some debugging checkpoints I'd be happy to run these again if it helps to locate the problem/bug.

from crest.

 avatar commented on June 4, 2024

The new 2.11 release (with xtb 6.3.3) appears to have fixed this problem, so closing this now.

from crest.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.