Difference between revisions of "Data Management"

From crtc.cs.odu.edu
Jump to: navigation, search
(CNF)
(Particle Trajectory Tracking with ML)
 
(31 intermediate revisions by 4 users not shown)
Line 1: Line 1:
 +
 +
== Next-generation imaging filters and mesh-based data representation for phase-space calculations in nuclear femtography ==
 +
 +
<gallery heights=300px widths=300px mode="packed-hover" caption="Tomographic pictures of the nucleon">
 +
File:Paraview_plot.png|350px|
 +
</gallery>
 +
 +
Tomographic and recently aquired and tessellated pictures of the nucleon as a result of this project. Namely,  the plots show a spatial distribution of up quarks as a function of proton's momentum fraction carried by those quarks.  Specifically, bX and bY are the spatial coordinates (in 1/GeV = 0.197 fm) defined in a plane perpendicular to the nucleon’s motion, x is the fraction of proton’s momentum and color denotes probability density for finding a quark at given (bX, bY, x). 
 +
 +
Plots produced by  Dr. Gagik Gavalian and Dr. Pawel Sznajder and tesselated by CRTC's Image-to-Mesh (I2M) conversion software deployed to Jefferson Lab last month.
 +
 +
'''For more data and information about this project follow this link: [[CNF]]'''
 +
 +
== Telescopic Approach ==
 +
<gallery heights=600px widths=600px mode="packed-hover" caption="The telescopic approach of CRTC lab">
 +
File:New_Telescopic_Approach.png|600px|
 +
</gallery>
 +
 +
 +
The project aims at investigating the design and implementation of multi-layered algorithmic and software framework for 3D tetrahedral parallel mesh generation. The framework is referred as the [[Telescopic Approach]]
 +
 +
== Exascale-Era Finite Element Mesh Generation  ==
 +
 
[[File:CDT3D_pipeline.png |800px]]
 
[[File:CDT3D_pipeline.png |800px]]
 +
 
== Boundary Recovery ==
 
== Boundary Recovery ==
  
Line 16: Line 40:
 
[[Runtime Systems]]
 
[[Runtime Systems]]
  
== CNF ==
+
== Particle Trajectory Tracking with ML ==
[[CNF Demo]]
+
[[CNF:Machine_Learning | Machine Learning for Track Classification and Prediction]]
[https://odu.box.com/s/ce6wc1ht1f96jx7lpafys1ue1piiyekc Shared box directory] (restricted access)
 
 
 
[https://git-community.cs.odu.edu/aangelop/PODM-Superbuild PODM Docker image source files] (restricted access)
 
 
 
[http://www.cs.odu.edu/crtc/cnf_project/index.html CNF Tools website]
 
  
[https://git-community.cs.odu.edu/ctsolakis/cnf_website CNF Tools webite build files] (restricted access)
+
[[CNF:ML_Data_Denoise | Machine Learning for Track Denoising]]

Latest revision as of 23:35, 3 November 2020

Next-generation imaging filters and mesh-based data representation for phase-space calculations in nuclear femtography

Tomographic and recently aquired and tessellated pictures of the nucleon as a result of this project. Namely, the plots show a spatial distribution of up quarks as a function of proton's momentum fraction carried by those quarks. Specifically, bX and bY are the spatial coordinates (in 1/GeV = 0.197 fm) defined in a plane perpendicular to the nucleon’s motion, x is the fraction of proton’s momentum and color denotes probability density for finding a quark at given (bX, bY, x).

Plots produced by Dr. Gagik Gavalian and Dr. Pawel Sznajder and tesselated by CRTC's Image-to-Mesh (I2M) conversion software deployed to Jefferson Lab last month.

For more data and information about this project follow this link: CNF

Telescopic Approach


The project aims at investigating the design and implementation of multi-layered algorithmic and software framework for 3D tetrahedral parallel mesh generation. The framework is referred as the Telescopic Approach

Exascale-Era Finite Element Mesh Generation

CDT3D pipeline.png

Boundary Recovery

Boundary Recovery

Isotropic Mesh Generation

Isotropic Mesh Generation

Anisotropic Mesh Generation

Anisotropic Mesh Generation

Runtime Systems

Runtime Systems

Particle Trajectory Tracking with ML

Machine Learning for Track Classification and Prediction

Machine Learning for Track Denoising