pyDOE 0.2.2

Design of experiments for Python

The pyDOE package is designed to help the scientist, engineer, statistician, etc., to construct appropriate experimental designs.

Capabilities

The package currently includes functions for creating designs for any number of factors:

• Factorial Designs

  1. General Full-Factorial (fullfact)

  2. 2-level Full-Factorial (ff2n)

  3. 2-level Fractional Factorial (fracfact)

  4. Plackett-Burman (pbdesign)

• Response-Surface Designs

  1. Box-Behnken (bbdesign)

  2. Central-Composite (ccdesign)

• Randomized Designs

  1. Latin-Hypercube (lhs)

See the package homepage for details on usage and other notes

Requirements

• NumPy

• SciPy

What’s New

In this release, some significant improvements and flexibility have been added to the latin-hypercube sampling function (see lhs documentation for details):

• The “criterion” keyword-argument has been added to allow for more advanced sampling techniques (default simply randomizes within the sampling intervals)

  • centered within sampling intervals

  • maximize the minimum distance between points

  • minimize the maximum correlation coefficient

Installation and download

See the package homepage for helpful hints relating to downloading and installing pyDOE.

Source Code

The latest, bleeding-edge but working code and documentation source are available on GitHub.

Contact

Any feedback, questions, bug reports, or success stores should be sent to the author. I’d love to hear from you!

License

This package is provided under two licenses:

1. The BSD License

2. Any other that the author approves (just ask!)

References

• Factorial designs

• Plackett-Burman designs

• Box-Behnken designs

• Central composite designs

• Latin-Hypercube designs