portelli/Grid
1
0
Fork 0
mirror of https://github.com/paboyle/Grid.git synced 2025-06-17 15:27:06 +01:00
Code Releases Activity

Imported changes from feature/gparity_HMC branch:

Browse Source 
Rework of WilsonFlow class
		Fixed logic error in smear method where the step index was initialized to 1 rather than 0, resulting in the logged output value of tau being too large by epsilon
		Previously smear_adaptive would maintain the current value of tau as a class member variable whereas smear would compute it separately; now both methods maintain the current value internally and it is updated by the evolve_step routines. Both evolve methods are now const.
		smear_adaptive now also maintains the current value of epsilon internally, allowing it to be a const method and also allowing the same class instance to be reused without needing to be reset
		Replaced the fixed evaluation of the plaquette energy density and plaquette topological charge during the smearing with a highly flexible general strategy where the user can add arbitrary measurements as functional objects that are evaluated at an arbitrary frequency
	        By default the same plaquette-based measurements are performed, but additional example functions are provided where the smearing is performed with different choices of measurement that are returned as an array for further processing
		Added a method to compute the energy density using the Cloverleaf approach which has smaller discretization errors
	Added a new tensor utility operation, copyLane, which allows for the copying of a single SIMD lane between two instances of the same tensor type but potentially different precisions
	To LocalCoherenceLanczos, added the option to compute the high/low eval of the fine operator on every restart to aid in tuning the Chebyshev
	Added Test_field_array_io which demonstrates and tests a single-file write of an arbitrary array of fields
	Added Test_evec_compression which generates evecs using Lanczos and attempts to compress them using the local coherence technique
	Added Test_compressed_lanczos_gparity which demonstrates the local coherence Lanczos for G-parity BCs
	Added HMC main programs for the 40ID and 48ID G-parity lattices
This commit is contained in:
Christopher Kelly
2022-07-01 14:10:59 -04:00
parent fd933420c6
commit 33e4a0caee
9 changed files with 3247 additions and 57 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
Grid/algorithms/iterative/LocalCoherenceLanczos.h
View File

@ -146,14 +146,21 @@ public:
LinearOperatorBase<FineField> &_Linop;
RealD _coarse_relax_tol;
std::vector<FineField> &_subspace;
int _largestEvalIdxForReport; //The convergence of the LCL is based on the evals of the coarse grid operator, not those of the underlying fine grid operator
//As a result we do not know what the eval range of the fine operator is until the very end, making tuning the Cheby bounds very difficult
//To work around this issue, every restart we separately reconstruct the fine operator eval for the lowest and highest evec and print these
//out alongside the evals of the coarse operator. To do so we need to know the index of the largest eval (i.e. Nstop-1)
//NOTE: If largestEvalIdxForReport=-1 (default) then this is not performed
ImplicitlyRestartedLanczosSmoothedTester(LinearFunction<CoarseField> &Poly,
OperatorFunction<FineField> &smoother,
LinearOperatorBase<FineField> &Linop,
std::vector<FineField> &subspace,
RealD coarse_relax_tol=5.0e3)
RealD coarse_relax_tol=5.0e3,
int largestEvalIdxForReport=-1)
: _smoother(smoother), _Linop(Linop), _Poly(Poly), _subspace(subspace),
_coarse_relax_tol(coarse_relax_tol)
_coarse_relax_tol(coarse_relax_tol), _largestEvalIdxForReport(largestEvalIdxForReport)
{ };
//evalMaxApprox: approximation of largest eval of the fine Chebyshev operator (suitably wrapped by block projection)
@ -179,6 +186,12 @@ public:
<<" |H B[i] - eval[i]B[i]|^2 / evalMaxApprox^2 " << std::setw(25) << vv
<<std::endl;
if(_largestEvalIdxForReport != -1 && (j==0 || j==_largestEvalIdxForReport)){
std::cout<<GridLogIRL << "Estimating true eval of fine grid operator for eval idx " << j << std::endl;
RealD tmp_eval;
ReconstructEval(j,eresid,B,tmp_eval,1.0); //don't use evalMaxApprox of coarse operator! (cf below)
}
int conv=0;
if( (vv<eresid*eresid) ) conv = 1;
return conv;
@ -409,7 +422,7 @@ public:
//////////////////////////////////////////////////////////////////////////////////////////////////
Chebyshev<FineField> ChebySmooth(cheby_smooth); //lower order Chebyshev of fine operator on fine grid used to smooth regenerated eigenvectors
ImplicitlyRestartedLanczosSmoothedTester<Fobj,CComplex,nbasis> ChebySmoothTester(ChebyOp,ChebySmooth,_FineOp,subspace,relax);
ImplicitlyRestartedLanczosSmoothedTester<Fobj,CComplex,nbasis> ChebySmoothTester(ChebyOp,ChebySmooth,_FineOp,subspace,relax,Nstop-1);
evals_coarse.resize(Nm);
evec_coarse.resize(Nm,_CoarseGrid);