Molcas Forum

Support and discussions for Molcas and OpenMolcas users and developers

You are not logged in.

Announcement

Welcome to the Molcas forum.

Please note: The forum's URL has changed. The new URL is: https://molcasforum.univie.ac.at. Please update your bookmarks!

You can choose an avatar and change the default style by going to "Profile" → "Personality" or "Display".

#1 2018-08-16 08:50:52

Sarah
Member
Registered: 2016-08-01
Posts: 40

CASPT2-HESSIAN--------Any suggestion?

Dear Molcas Team,

When calculating 'Analytical Hessian', you get a "filename.UnSym" file, where you can extract the hessian data from. But I don't know where to find the hessian data when I do a 'CASPT2 Numerical Hessian' calculations.

Here is my input:

&gateway
 coord
 2
 Angstrom
 H 0.0 0.0 -0.37
 H 0.0 0.0  0.37
 Basis=6-31g
 group=C1
>>>  Export  MOLCAS_MAXITER=800
>>> Do while

&SEWARD

>>>  IF  (  ITER  =  1  )  THEN
  &SCF
  TITLE=CASPT2HESSIAN
>>> EndDo
&RASSCF
  Title
  charge   = 0
  spin     = 1
  nactel   = 2 0 0
  inactive = 0
  ras2     = 3
&CASPT2

&ALASKA
&SLAPAF
Numerical Hessian
>>>  COPY  $Project.RunFile  $CurrDir/$Project.ForceConstant.ANO

Thanks

Last edited by Sarah (2018-08-16 14:52:11)

Offline

#2 2018-08-17 15:31:40

Ignacio
Administrator
From: Uppsala
Registered: 2015-11-03
Posts: 1,085

Re: CASPT2-HESSIAN--------Any suggestion?

In the latest OpenMolcas, if you use HDF5, you can find a $Project.slapaf.h5 that includes the Hessian.

Offline

#3 2018-08-17 16:35:52

Sarah
Member
Registered: 2016-08-01
Posts: 40

Re: CASPT2-HESSIAN--------Any suggestion?

Thank you, Ignacio,

how can I use OpenMolcas? Should it be downloaded and installed as a separate program?

Last edited by Sarah (2018-08-17 16:36:11)

Offline

#4 2018-08-17 20:52:44

Ignacio
Administrator
From: Uppsala
Registered: 2015-11-03
Posts: 1,085

Re: CASPT2-HESSIAN--------Any suggestion?

Yes.

Offline

#5 2018-08-20 15:59:55

Sarah
Member
Registered: 2016-08-01
Posts: 40

Re: CASPT2-HESSIAN--------Any suggestion?

Hi Ignacio,

I am in the last step of installing OpenMolcas, I ran "pymolcas verify" around one hour ago, I pressed enter meanwhile because it was running very slowly. It is what I have got so far:

.
.
.
The installation of pymolcas was successful
[100%] Built target pymolcas
[....@.....build]$ pymolcas verify


Running test standard: 000 ... (0%)
Running test standard: 015 ... (3%)
Running test standard: 017 ... (4%)

Running test standard: 020 ... (5%)

Is that normal? I tried to run an input in the test folder (test/hdf5), but it complains: "/work/OpenMolcas/OpenMolcas" is not a valid MOLCAS installation.

In the second try, I ran the first input in this post by the following command: pymolcas t.inp and here is my output:

           *************************************************
           * pymolcas version py2.03                       *
           *   build 7c560574aa310260080e75d42b56265c      *
           *   (after the EMIL interpreter by V. Veryazov) *
           *************************************************


   -----------------------------------------------
  |
  |           Project: t
  |    Submitted from: /work/sbazzi/OpenMolcas/try
  |      Scratch area: /tmp/t
  |   Save outputs to: /work/sbazzi/OpenMolcas/try
  |            Molcas: /work/bin/molcas_dir/8.0
  |
  | Scratch area is empty
  |
  |    MOLCAS_DRIVER = /home/sophia/bin/pymolcas
  |    MOLCAS_NPROCS = 1
  |    MOLCAS_SOURCE = /work/bin/molcas_dir/8.0
  | MOLCAS_STRUCTURE = 0
  |
   -----------------------------------------------

++ ---------   Input file   ---------

&GATEWAY
 coord
 2
 Angstrom
 H 0.0 0.0 -0.37
 H 0.0 0.0  0.37
 Basis=6-31g
 group=C1

&SEWARD

&SCF

&RASSCF
  Title
  charge   = 0
  spin     = 1
  nactel   = 2 0 0
  inactive = 0
  ras2     = 3

&CASPT2

&ALASKA

&SLAPAF
Numerical Hessian

-- ----------------------------------

Aborting...
Traceback (most recent call last):
  File "/home/sophia/bin/pymolcas", line 449, in <module>
    sys.exit(main(os.path.realpath(f)))
  File "<string>", line 206, in main
  File "<string>", line 686, in auto
  File "<string>", line 190, in run
  File "<string>", line 342, in run
  File "<string>", line 745, in run_module
  File "<string>", line 1197, in run
  File "<string>", line 1163, in _make_links
  File "<string>", line 942, in parse_prgm
  File "<string>", line 72, in utf8_open
FileNotFoundError: [Errno 2] No such file or directory: '/work/bin/molcas_dir/8.0/data/inporb_files.prgm'

Thanks

Last edited by Sarah (2018-08-20 16:08:02)

Offline

#6 2018-08-20 16:08:19

Ignacio
Administrator
From: Uppsala
Registered: 2015-11-03
Posts: 1,085

Re: CASPT2-HESSIAN--------Any suggestion?

Depending on your machine and settings, the verification can take several hours, as indicated in the Quick start guide. If it only reached test 020 in one hour, it's being extremely slow, which may be due to I/O if the disk where you are running is mounted in a network (likely if you are running on your home directory in a cluster). For running inputs note that you have to use "pymolcas" and not "molcas".

Offline

#7 2018-08-20 16:46:37

Ignacio
Administrator
From: Uppsala
Registered: 2015-11-03
Posts: 1,085

Re: CASPT2-HESSIAN--------Any suggestion?

You have to define MOLCAS to point to the version you want to use, you can see it says "Molcas: /work/bin/molcas_dir/8.0"

Offline

#8 2018-08-20 17:25:18

Sarah
Member
Registered: 2016-08-01
Posts: 40

Re: CASPT2-HESSIAN--------Any suggestion?

Thanks a lot, Ignacio, I modified my .bashrc with the right molcas version and I could run a test input successfully. However, I don't know what you mean by "if you use HDF5". How can I use HDF5?

Offline

#9 2018-08-21 08:45:35

Ignacio
Administrator
From: Uppsala
Registered: 2015-11-03
Posts: 1,085

Re: CASPT2-HESSIAN--------Any suggestion?

Just make sure HDF5 is enabled when you run cmake. If it's not, you probably need to install/activate HDF5 libraries in your system. Once it's compiled, you should get *.h5 files normally in your calculations.

Offline

#10 2018-08-21 09:14:02

Sarah
Member
Registered: 2016-08-01
Posts: 40

Re: CASPT2-HESSIAN--------Any suggestion?

running cmake gave me the following:

Configuring HDF5 support:
-- Found HDF5: /home/sophia/anaconda3/lib/libhdf5.so;/usr/lib64/librt.so;/usr/lib64/libpthread.so;/home/sophia/anaconda3/lib/libz.so;/usr/lib64/libdl.so;/usr/lib64/libm.so 
-- HDF5_INCLUDE_PATH: /home/sophia/anaconda3/include
-- HDF5_C_LIBRARIES: /home/sophia/anaconda3/lib/libhdf5.so;/usr/lib64/librt.so;/usr/lib64/libpthread.so;/home/sophia/anaconda3/lib/libz.so;/usr/lib64/libdl.so;/usr/lib64/libm.so

Configuring linear algebra libraries:
-- Using Molcas' LAPACK+BLAS libraries (SLOW!)
-- LINALG_LIBRARIES: lapack;blas
Gromacs interface DISABLED
BLOCK interface DISABLED
CHEMPS2 interface DISABLED
-- MSYM support DISABLED
-- DMRG support DISABLED
-- HDF5 QCMaquis interface DISABLED
NECI interface DISABLED
EFP interface DISABLED
GEN1INT support DISABLED
-- libwfa support DISABLED
-- NEVPT2 support DISABLED
External GRID_IT DISABLED
Configuring runtime environment settings:
-- DEFMOLCASMEM:  2048
-- DEFMOLCASDISK: 20000
-- RUNSCRIPT:    $program $input
-- RUNBINARY:    $program
-- RUNBINARYSER: $program
Build type: Release
-- C compiler: /usr/bin/cc
-- C compiler flags:  -std=gnu99 -O2
-- Fortran compiler: /usr/bin/f95
-- Fortran compiler flags:  -cpp -fdefault-integer-8 -O2
-- Definitions: _MOLCAS_;_I8_;_LINUX_
-- Debug definitions:
-- Found PythonInterp: /home/sophia/anaconda3/bin/python3 (found suitable version "3.6.5", minimum required is "3.0")
-- pymolcas: added to targets
Definitions: -D_MOLCAS_;-D_I8_;-D_LINUX_
Configuring documentation
-- LaTeX compiler: no latexmk available, documentation disabled
Install directory: /opt/molcas
-- Configuring done
-- Generating done
-- Build files have been written to: /work/sbazzi/OpenMolcas/build

And running pymolcas verify resulted in the following so far:

standard:057 Failed! (alaska)       
standard:069 Failed! (alaska)       
standard:072 Skipped!               
standard:078 Failed! (numerical_gradient)
standard:079 Failed! (numerical_gradient)
standard:080 Failed! (numerical_gradient)
Running test additional: 100 ... (23%)

Offline

Board footer

Powered by FluxBB 1.5.11

Last refresh: Today 21:52:47