q2mm icon indicating copy to clipboard operation
q2mm copied to clipboard

Request for logging of intermediate data

Open ericchansen opened this issue 10 years ago • 3 comments

Older versions of the code generated tables of data in files par.one, par.ref, and par.tot. A description of the files can be found in Per-Ola Norrby and Tommy Liljefors, J. Comput. Chem. 1998, 19, 1146-1166. It would be nice to still have access to these files or perhaps something similar.

ericchansen avatar Aug 26 '15 14:08 ericchansen

There have been a lot of updates in regards to this. For the desired level of verbosity, try adjusting the logging levels in the constants module. For more information on logging levels, please refer to the documentation.

I'm going to add an option to write the Jacobian matrix to a file. Once this has been completed, I'll consider this issue closed.

ericchansen avatar Mar 01 '16 15:03 ericchansen

I think once the Jacobian read and write functionality is added, this issue can be closed.

ericchansen avatar Apr 22 '16 04:04 ericchansen

Commit cb36b4c1bece4723a has helped move this issue forward significantly. This enabled one to restart the gradient optimization from the parameter central differentiation file.

I think all that is needed now is the option to print or write the Jacobian.

This probably will necessitate that gradient be runnable from the command line with argument handling.

ericchansen avatar Jun 15 '16 15:06 ericchansen