TSPTW using Parallel Stable NRPA
Description
Implementation of a parallel Stable NRPA to solve the TSPTW as describe by Tristan Cazenave
in Stabilized Nested Rollout Policy Adaptation.
The implementation uses an actors model to handle the concurrent computation of the NRPA leaves (playout) and the
concurrent runs of different trees.
Related paper (in spanish)
Prerequisites
You need to download the repository and place it in the src folder inside your GOPATH folder.
Build
go build main.go
Run
./main -cpuprof=nrpa.prof -runs=5 -all (args...)
Program arguments
Command |
Type |
Description |
Default |
cpuprof |
string |
Filename of the generated report |
|
file |
string |
File name of a test case |
rc_201.1.txt |
time |
int |
Execution timeout in sec |
10000 |
factor |
int |
Tree stabilization factor |
10 |
iter |
int |
Iterations in the next level (number of children per node) |
10 |
levels |
int |
Tree levels |
7 |
runs |
int |
Number of trees to run |
4 |
all |
bool |
read all case files |
false |
nactors |
int |
number of actors that computes the NRPA tree |
same as runs |
pactors |
int |
number of actors that computes the leaves. |
same as factor |
Author
License
This project is licensed under the Apache-2.0 License - see the LICENSE file for details