Skip to content

lxAltria/Multiprecision-data-refactoring

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

62 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Multi-precision data refactoring framework

Project: ESAMR (Enabling Scalable Analysis using Multi-precision Refactoring)
Sponsor: ORNL LDRD
Authors: Xin Liang, Jieyang Chen, Lipeng Wan, Ben Whitney, Qian Gong
Supervisors: Scott Klasky, Rick Archibald

Installation

git clone https://github.com/lxAltria/Multiprecision-data-refactoring.git
cd Multiprecision-data-refactoring
./build_script.sh

Usage

cd build
mkdir -p refactored_data
Refactor: ./test/test_refactor $data_file $num_level $num_bitplanes $num_dims $dim0 $dim1 $dim2
./test/test_refactor ../external/SZ3/data/Uf48.bin.dat 4 32 3 100 500 500
Retrieval: ./test/test_retrieval $data_file $error_mode $error $s
./test/test_reconstructor ../external/SZ3/data/Uf48.bin.dat 0 1.0 0

Notes and Parameters

During refactoring, the location of refactored data is hardcoded to "refactored_data/" directory under current directory. Need to create the directory before writing.
data_file: path to input date file.
num_levels: number of target decomposition levels.
num_bitplanes: number of bitplanes for each level.
num_dims: number of dimensions.
Option: options of encoder/decomposer/retrieval etc. are changeable, but not supported in commandline for now (see these components in different folders of include and alter the options in test/test_refactor.cpp and test/test_reconstruct.cpp)
error mode: error metric during retreival (see include/error_est.hpp)
0: max error, i.e. L-infty
1: squared error, i.e. L-2

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published