Skip to content

A research project supported by DIMACS REU. Built a two-level random forests classifier that aims to automate 3D printing quality inspection. Paper published on IEEE Sensor Letter.

Notifications You must be signed in to change notification settings

328Machunyi/DIMACS-REU-2018

 
 

Repository files navigation

In this folder, you’ll find

  • quality_metrics.Rmd: R code that defines profile residual.
  • modeling.Rmd: R code of the classifiers. It includes K-means clustering, PCA and Random Forest.
  • measurements folder: all the measurement matrices
  • umpca folder: umpca related matlab code and reconstruction
  • qualitydf.csv: the quality df exported from quality_metrics.Rmd

Within umpca:

  • fea_extraction_umpca.m: matlab code that calculates features extracted from high VR data
  • matrices folder: all VR measurement matrices. They are exported from modeling.Rmd
  • mexeig.m, UMPCA.m, tensor_toolbox: UMPCA toolbox
  • reconstruction: reconstructed matrices exported from matlab code and R code that visualizes the reconstruction
  • newfea_test.csv, newfea_train.csv: features extracted

About

A research project supported by DIMACS REU. Built a two-level random forests classifier that aims to automate 3D printing quality inspection. Paper published on IEEE Sensor Letter.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • HTML 90.9%
  • MATLAB 9.1%