![]() ![]() ![]() Edit the 'nSig' variable in the above script to reproduce any one of the subpanels of Fig.10 and Fig.11 in the paper.This is because it generates and analyzes 1000 simulated data files. Assuming that there are 4 parallel workers available, the runtime of this script is about 2.7 hours on a 3 GHz 8-Core Intel Xeon E5. The script validate_Fig10.m reproduces the panel for benchmark function f_2 in Fig.In addition, each function has associated usage instructions that can be accessed using Matlab's "help" command.Each test_*.m script also serves as an example of how to use the respective function.For reference, test_shps.m, which tests the main function shps.m, takes about 9 sec to complete on a 3 GHz 8-Core Intel Xeon E5 on one data realization with 260 samples, one PSO run, and other algorithm settings as given in the paper.Hence, use the latter to first generate some data. It runs SHAPES on data generated by test_gendataBFsig.m. An example is provided in test_shps.m.Return to the SHAPES folder and run the test_*.m scripts in the SHAPES folder to test your installation using the above data files.Enter the SHAPES / DATA folder and run the test_gendataBFsig.m script to generate sample files containing simulated data. ![]() Alternatively, permanently add the paths to your Matlab search path list using addpath/savepath or through the GUI interface accessible from the "Set Path" button. Run pathset.m (required only once at the start of a Matlab session).Edit the pathset.m file in the SHAPES repository to provide the path to the SDMBIGDAT19 directory in your local directory tree.Clone the repository containing Particle Swarm Optimization (PSO) codes: SDMBIGDAT19.The codes are in Matlab and have been tested under "Adaptive spline fitting with particle swarm optimization." arXiv preprint arXiv:1907.12160 (2019). Codes for the SHAPES (Swarm Heuristics based Adaptive and PenalizedĮstimation of Splines) adaptive spline curve fitting method described in: ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. Archives
March 2023
Categories |