TCE CCSD print "t2" output

From NWChem

Viewed 2198 times, With a total of 3 Posts
Jump to: navigation, search

Clicked A Few Times
Threads 18
Posts 44
I am running a test CCSD/cc-pVTZ job on a neutral Li atom, with an ROHF reference. The job completes fine, but I have a question about the output generated with 'print "t2"' in the tce block.

I see two types of output lines that look important. One is of the form

#1 #2 #3 #4 has been assigned to node X

which looks from the source like specification of the double excitation (#s 1 and 2 particle orbitals, 3 and 4 hole orbitals).

The second line type is more cryptic. An example,

    1    1 -0.8258539073E-04

The second integer in the test run is always "1", but the first runs to 289. I would guess the floats are amplitudes, but what are the integers?

Also, the two line types are interspersed with each other; sometimes the first type of line occurs without the second type, and other times there follows 1+ lines of the second type:
...
 t2
-------------------------

  4  6  3  3 has been assigned to node   0

  4  7  1  1 has been assigned to node   0

  4  7  1  2 has been assigned to node   0

  4  7  1  3 has been assigned to node   0

 t2
-------------------------
    1    1 -0.8258539073E-04


Is this intentional (for example, a 3-->4 + 3-->6 excitation is nonsensical because there is only 1 electron in block 3), or are multiple MPI tasks printing on top of each other?

Thanks

  • Karol Forum:Admin, Forum:Mod, NWChemDeveloper, bureaucrat, sysop
    Profile
    Send PM
Clicked A Few Times
Threads 1
Posts 36
Dear Chris,
using print option in the tce input is not he best way of printing cluster amplitudes.
This is usually used fro the debugging. Instead you can use the printtol option, which
prints all cluster amplitudes with corresponding absolute value grater than the value of
printtol (the default value of printtol is 0.1). You can change this value in your input
by adding:

set tce: printtol 0.01

before the "task tce energy" statement.
Be careful, sometimes setting printtol too small will result is huge output files.
Best,
Karol

Clicked A Few Times
Threads 18
Posts 44
Karol--Thanks so much, I'll give it a try.
- Chris

Gets Around
Threads 3
Posts 85
Quote:Cchang Nov 27th 12:07 am
I am running a test CCSD/cc-pVTZ job on a neutral Li atom, with an ROHF reference. The job completes fine, but I have a question about the output generated with 'print "t2"' in the tce block.

I see two types of output lines that look important. One is of the form

#1 #2 #3 #4 has been assigned to node X

which looks from the source like specification of the double excitation (#s 1 and 2 particle orbitals, 3 and 4 hole orbitals).

The second line type is more cryptic. An example,

    1    1 -0.8258539073E-04

The second integer in the test run is always "1", but the first runs to 289. I would guess the floats are amplitudes, but what are the integers?

Also, the two line types are interspersed with each other; sometimes the first type of line occurs without the second type, and other times there follows 1+ lines of the second type:
...
 t2
-------------------------

  4  6  3  3 has been assigned to node   0

  4  7  1  1 has been assigned to node   0

  4  7  1  2 has been assigned to node   0

  4  7  1  3 has been assigned to node   0

 t2
-------------------------
    1    1 -0.8258539073E-04


Is this intentional (for example, a 3-->4 + 3-->6 excitation is nonsensical because there is only 1 electron in block 3), or are multiple MPI tasks printing on top of each other?

Thanks


It is possible that multiple MPI processes are writing to stdout in such a way as to produce a confusing result.

In general, debug printout like this is only useful to developers. I discourage trying to use it unless you are familiar with the corresponding source code.

In any case, Karol has given the right answer already. In the event that you need to change the output format, I can tell you how to do that.

Jeff


Forum >> NWChem's corner >> Running NWChem



Who's here now Members 0 Guests 1 Bots/Crawler 0


AWC's: 2.5.10 MediaWiki - Stand Alone Forum Extension
Forum theme style by: AWC