qmcpack_input_generator

Takes a file from quantum package that is ready for conversion (i.e. save_for_qmcpack has been used) and generates the *wfs.xml, *ptcl.xml, DMC folder, and Optimization folder (defaulting to adding only the 1 and 2 body Jastrows).

Uses

./setup_QMC_calculation.py setup --filename=<filename.ext> --method=QP [--noJastrow=<True,False>] [--3BodyJ=<True,False>] [--reoptimizeCoeffs=<True,False>] [--coeffCutoffValue=]

                Filename: the file that is ready for conversion to qmcpack
                Method: at the moment only quantum package input can be used, but it would be relatively easy to modify the setupMethods.py file if another input is desired.
                --noJastrow= default is false. Set to true if you do NOT want 1,2, or 3 body jastrow
                --3BodyJ   =default is False. Set to false if you only want 1 and 2 body Jastrows.
                --reoptimizeCoeffs = default is False. Set to true if you want to reoptimize multideterminant coefficients
	    --coeffCutoffValue = default is 0.01. You can change this value to keep the determinants with coefficients greater or equal to this value

Output

This script can be called from the directory with the input file and generates all needed input for QMC.
The *wfs.xml and *ptcl.xml files will be generated in a subdirectory called JASTROW_NDET(_reopt). Note: The 
If any optimization will take place the Optimization folder will be created with the Opt.xml file (The Opt.xml file is either misc/Opt_AE.xml or a version of misc/Opt_PP.xml with the proper pseudopotential Hamiltonian). 
A DMC folder will be created with the needed DMC.xml file (The DMC.xml file is either misc/DMC_AE.xml or a version of misc/DMC_PP.xml with the proper pseudopotential Hamiltonian). 
The Hamiltonians in each are created based on whether a pseudopotential was used. 
If it's an all-electron calculation then a Cusp Correction folder will be created.

Example of use

   cd myTest
   pwd 
       ~/myTest
   ls
       thisSystem.dump

##Running within the generated directories using the generated files

   ~/qmcpack_input_generator/setup_QMC_calculation.py setup --filename=thisSystem.dump --method=QP --3BodyJ=True
   
'''We have received all needed info
	do_psuedo =  False
	The file is from quantum package
	The input files will be place in  Jastrow123_MultiDet
	Beginning conversion
	Rank =    0  Free Memory = 13699 MB
	Finished Conversion
	Multi reference system
	This is an all electron calculation so the Cusp correction is being added
	setupMethod.py is done
	Everything is done
	'''
ls
     thisSystem.dump     conversion.out     Jastrow123_Multi
ls Jastrow123_Multi/
     thisSystem.wfs.xml    thisSystem.ptcl.xml    CuspCorrection  DMC/   Optimization/

ls Jastrow123_Multi/DMC/
     DMC.xml
     
ls Jastrow123_Multi/Optimization/
     Opt.xml

qmcpack optimizer files

./optimize_system.py --optFile optFile.xml --wfsFile wfsFile.xml --optType {} --seriesNum {1,2,...,999}

optFile: The name of optimization file that will be used ( It should already exist in the directory)
wfsFile: The name of wavefunction file that will be used 
optType: 
	 12 -> optimize 1 and 2 body Jastrows only
	 RC -> include the coefficient reoptimization
	 3B -> include the 3 body Jastrow in the optimization
	 Fin -> check to see if the system is ready for DMC

serieNum: The series number to use for the next round of optimization or DMC (any number from 1 to 999)


It also outputs "opt_run_info.dat" to help keep track of which optimization type was done for which group of files

What the script does internally

1) It goes through the input arguments and checks that a valid optimization and wavefunction file are given. It then checks that an optimization type is given.
2) It checks which *opt.xml files have been output and sets the restart number as the last *.opt.xml +1
3) It opens the optimization file and checks where the group started and then sets the series number as the restart number. It also pulls the project id for later use.
4) If no optimization run has been done, it will generate the "opt_run_info.dat" file. If only one run has been done, it will initialize the data. If mulitple runs have been done, then it will pull information from the file (using the run info to create a wavefunction backup, which the user can delete if needed).
6) The wavefunction file is backed up, and if there has been an optimization run the user supplied wavefunction will replace it
7) It looks at the wavefunction and determines which optimizations have been done ( for output to "opt_run_info.dat")
8) If the wavefunction needs to be modifies (i.e. the first step in optimization is needed or we are moving to the next optimization step), then it is changed based on user input (i.e. the value of optType)
9) If we have overwritten the wavefunction file with a series optimziation file, then we add a comment into the main wavefunction file which specifies the series number.
10) If the opt_run_info data exists, then it is output to the file 
11) Finally, if no errors occured, then the modified optimization file replaces the original optimization file and the modified wavefunction file replaces the original wavefunction file. (If an error occurs, then the modified files may exist as "Filename.ext.tmp")

output files

opt_run_info.dat: type of optimization, the run number (for saving wavefunction files), series number used for the wavefunction,series number for start of the group