Skip to content

Utilities for making eigenvector, elemental and perambulator in the framework of the distillation method in lattice QCD.

License

Notifications You must be signed in to change notification settings

IHEP-LQCD/EasyDistillation

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

EasyDistillation

EasyDistillation is a Python package that integrates the procedure of distillation method as proposed first by ref.[X]. This package makes up of the generation of laplacian eigenvectors, perambulators, elementals, and automatic contraction, with multi-backend support both on CPU / GPU. It features a flexible and efficient designed for optimal performance with the Generator class. This package is desigened to simplify the complexities associated with the distillation process, make it accessible and efficient computationally for physical calculations using the distillation method.

File IO

Using latticce.preset to initialize metadata of input file. For example, file data GaugeField / Eigenvector / Elemental / Perambulator. and file type BinaryFile / IldgFile, numpy NdarrayFile, etc.

Generate Laplacian eigenvectors

Refer to EigenvectorGenerator in tests\test_eigenvector.py, the Laplacian eigensolover support either Python LinearOperator eigensolver with scipy on cpu / cupyx.sparse on GPU, or PyQuda + QUDA library. The gauge smeaing also depend on PyQuda.

Elemental

Refer to class ElementalGenerator in tests\test_elemental.py, as input eigenvector data, It generate up to n-deriv ($n<3$) elemental data sequentally.

Perambulator

Refer to class PerambulatorGenerator in tests/test_perambulator_mpi.py with PyQuda.

Contraction

Refer to example/gen_two_particle_corr_mom.py.

About

Utilities for making eigenvector, elemental and perambulator in the framework of the distillation method in lattice QCD.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published