Rank: 170 based on 513 downloads (last 30 days) and 15 files submitted
photo

Travis Wiens

E-mail
Company/University
University of Saskatchewan

Personal Profile:

http://blog.nutaksas.com

Professional Interests:
mechanical engineering, dsp, embedded, controls, hydraulics, fluid power

 

Watch this Author's files

 

Files Posted by Travis View all
Updated   File Tags Downloads
(last 30 days)
Comments Rating
24 Jun 2013 Screenshot Brewing Calculators A number of functions for brewers. Author: Travis Wiens beer, wine, brewing, homebrew, refractometer, specific gravity 7 0
18 Apr 2013 Screenshot Graph Picker Helps digitize graph data from an image. Author: Travis Wiens gui, digitizing, data import, image 22 1
  • 5.0
5.0 | 1 rating
16 Apr 2013 Screenshot Hydraulic System Comparison Gui Visualizes power losses and utilization for four common types of fluid power systems. Author: Travis Wiens mechanical, engineering, gui, hydraulic, fluid power, efficiency 11 0
11 Feb 2010 Screenshot Radial Basis Function Network Simulates and trains Gaussian and polyharmonic spline radial basis function networks. Author: Travis Wiens radial basis function, radial basis function..., rbf, integral, line integral, polyharmonic spline 120 4
  • 5.0
5.0 | 2 ratings
03 Feb 2010 Screenshot Linear Regression with Errors in X and Y Calculates slope and intercept for linear regression of data with errors in X and Y. Author: Travis Wiens mathematics, measurement, statistics, interpolation, modeling, linear regression 88 3
  • 5.0
5.0 | 3 ratings
Comments and Ratings by Travis View all
Updated File Comments Rating
22 Sep 2014 Linear Regression with Errors in X and Y Calculates slope and intercept for linear regression of data with errors in X and Y. Author: Travis Wiens

Rainer Boegle: This method is only for a single input.

01 Nov 2009 Complex Optimization of a Recurrent Neural Network Shows how to use the complex method to optimize a black-box neural network model of a load-sensing h Author: Travis Wiens

Note that the "complex method" package is now available at http://www.mathworks.com/matlabcentral/fileexchange/25428-complex-method-of-optimization

Comments and Ratings on Travis' Files View all
Updated File Comment by Comments Rating
13 Oct 2014 Day Length Calculates the length of daylight for any day and latitude Author: Travis Wiens Greene, Chad

This is a helpful function; thanks for sharing.

The input day format is somewhat non-user-friendly. The function header describes the input date as the "day of the year, counted starting with the day of the December solstice in the first year of a Great Year." To maintain Travis' functionality and get this function to work with Matlab's datenum format, add this to the beginning of the function:

if nargin>0 && any(Day>366)
[y,~,~] = datevec(Day);
Day = 11+datenum(Day)-datenum(y,1,1);
end

The above assumes that if any input date exceeds 366, it must be in datenum format. Then the function *approximates* day length assuming a static December solstice. Results will not be exact on the daily timescale, but should be sufficient to put multi-year data in datenum format in context with solar cycles.

08 Oct 2014 Linear Regression with Errors in X and Y Calculates slope and intercept for linear regression of data with errors in X and Y. Author: Travis Wiens Felix

I really like the code but is it possible to force the linear regression to go through the origin, i.e. a=0?

22 Sep 2014 Linear Regression with Errors in X and Y Calculates slope and intercept for linear regression of data with errors in X and Y. Author: Travis Wiens Wiens, Travis

Rainer Boegle: This method is only for a single input.

20 Sep 2014 Linear Regression with Errors in X and Y Calculates slope and intercept for linear regression of data with errors in X and Y. Author: Travis Wiens Boegle, Rainer

Can I use multiple regression and
can I omit the offset in the model?
E.g. Y is explained by two betas corresponding with two different x (of same type of measurement, i.e. same error), but no offset is fitted?

Y = X*beta + error
where X = [x1; x2]

27 Feb 2014 Thin Plate Spline Network with Radiohead Example Thin plate splines can be used to interpolate nonuniform data, such as the Radiohead dataset, which Author: Travis Wiens Verveld, Mark

Thanks for sharing, I like the examples.

Some constructive criticism:
You should really consider vectorizing this. In Matlab, whenever possible, avoid loops.

I just got a 66 fold reduction in computation time on the Radiohead example with N_t 3000 and N_grid 250. (From 626 s to 9.45 s)

Also, it is easy to generalize TPS to m-dimensional values in an n-dimensional domain. That is what makes the method so powerful IMO.

Contact us