Rank: 63960 based on 0 downloads (last 30 days) and 0 file submitted
photo

John

E-mail

Personal Profile:
Professional Interests:

 

Watch this Author's files

 

Comments and Ratings by John View all
Updated File Comments Rating
13 Sep 2012 MTIMESX - Fast Matrix Multiply with Multi-Dimensional Support Beats MATLAB 300% - 400% in some cases ... really! Author: James Tursa

I like Michael's reasoning and I think he is right about the 'double' result. So in general I think that the double should be the default. However, given that i enjoy being efficient with data, I would appreciate a 'logical' option.

11 Sep 2012 MTIMESX - Fast Matrix Multiply with Multi-Dimensional Support Beats MATLAB 300% - 400% in some cases ... really! Author: James Tursa

Personally, I do not understand why it should be a 'double' result. So, my vote goes towards a logical result.

However, would it be possible to add this as an option? (e.g.: mtimesx(a,b,'logical') will produce a logical result, if a and b are logical). This way the advanced user can be efficient with data, whereas the less experienced user will still get results that they are familiar with....

28 Aug 2012 MTIMESX - Fast Matrix Multiply with Multi-Dimensional Support Beats MATLAB 300% - 400% in some cases ... really! Author: James Tursa

Excellent code!

I experienced the same 'error' as Michael Völker, so I think automatically converting the logical/sparse data to double might be a good idea as an enhancement!

Nevertheless, great work! Saved me a lot of time/coding.

Contact us