Skip to content

Latest commit

 

History

History
49 lines (39 loc) · 1.46 KB

README.md

File metadata and controls

49 lines (39 loc) · 1.46 KB
Raw

The RAPIDS cuGraph library is a collection of GPU accelerated graph algorithms that process data found in GPU DataFrames.

Research: Infectious Disease Modeling, Monsoon 2020
Guided by: Prof. Kishore Kothapalli, Prof. Sriram Pemmaraju

See Colaboratory for notebook, but it takes about 15min to setup Rapid's cuGraph onto it, before starting. BlazingSQL has cuGraph already setup onto it, but you need to write your own notebook (it has no share feature). With BlazingSQL you need to copy data as well and main.py into its environment.


## OUTPUT
## Small graph: 5 vertices, 5 edges
Contact Graph:    (5, 2) (5, 3)
Infectable Graph: (5, 2) (2, 3)
Path graph:       (5, 2) (3, 3)
   src  dst  wt
1    0    2  10
2    3    0   9
   distance  vertex  predecessor
0       9.0       3            0
1      10.0       2            0
2       0.0       0           -1
## OUTPUT
## Large graph: 50000 vertices, 249272 edges
Contact Graph:    (50000, 2) (249272, 3)
Infectable Graph: (50000, 2) (112381, 3)
Path graph:       (50000, 2) (48619, 3)

references