Gradient-Boosted Trees Model used for Concrete Mixture Design Instance in ArXiv:1803.00952

Miten Mistry, Dimitrios Letsios, Gerhard Krennrich, Robert M. Lee and Ruth Misener

Imperial College London, UK

BASF SE, Germany

---

Overview

This repository contains the concrete mixture design gradient-boosted tree instance used in the numerical tests of:

• Mistry M, Letsios D, Krennrich G, Lee RM, and Misener R. 2018. Mixed-integer convex nonlinear optimization with gradient-boosted trees embedded. ArXiv eprints; 1803.00952.

Files

• trained_instance.RData: the trained instance using the parameters listed under Training Description below
• trained_instance.tree: the trained instance printed using pretty.gbm.tree.
• labels.dat: labels for variables
• lower.dat: vector of lower bounds
• means.dat: mean vector
• pca_load.dat: PCA loading vectors
• stddev.dat: vector of standard deviations
• upper.dat: vector of upper bounds

Training Description

• Data source: concrete compressive strength data set on UCI machine learning repository (Yeh, 1998; Dheeru and Karra Taniskidou, 2017).
• Gradient-boosted tree instance trained using:
  • R version 3.4.0
  • gbm version 2.1.3
  • caret version 6.0
  • caret trainControl parameters:
    • method: cv
    • number: 6
  • caret tuneGrid parameters:
    • n.trees: seq(1000, 8000, 100)
    • interaction.depth: c(2,4,6,8)
    • shrinkage: c(0.1, 0.05, 0.01)
    • n.minobsinnode: c(10)
  • caret metric parameter: RMSE
  • caret distribution: gaussian

References

• Dheeru D, Karra Taniskidou E. 2017. UCI machine learning repository. URL: http://archive.ics.uci.edu/ml/index.php.
• Yeh IC. 1998. Modeling of strength of high-performance concrete using artificial neural networks. Cement Concrete Research. 28(12):1797--1808.