DOE

utils for design DOE

Latin Hypercube Sampling

• statistical method for generating a near-random values from a multidimentional distribution
• LHS was designed by McKay in 1979
  • 
• Latin Square
  • There is only one sample in each row and each column
• Algorithm overview
  • When sampling a function of N variables
  • the range of each variable is devided into M
    • M is equally probable intervals
  • M sample points are then placed to satisfy the Latin Hypercube requirements
    • this force the number of divisions
    • M is equal for each variables
  • main advantage is "independence"
    • this sampling scheme does not require more points for more dimensions
  • another is that random sampling can be taken one at a time, remembering which samples were taken so far
• 

pyDOE

• this repository utilize