Re: Contribution of mapminmax equivalent function implementation

[siko1056]

Georgios Kousiouris wrote
> Hi all,
> 
> I have been working on a mapminmax function implementation for Octave 
> that i would like to contribute to the Octave community. It is nearly 
> done, only some details are missing such as help info etc (or 
> potentially other formatting aspects that are needed and i am not aware
> of).
> 
> The implementation tries to follow the Matlab defined interface of 
> mapminmax and to a large extent it covers it, plus adding some helpful 
> optional features that i have found to be important (such as adding an 
> offset to the denormalization phase for a matrix in order to denormalize 
> e.g. only the outputs).
> 
> What are the next steps in this process? Should i upload the function 
> somewhere, share it in the mailing list, some one needs to review it for 
> correctness etc?

Hello George,

Thank you for considering a contribution to Octave. New contribution should
be submitted to the patch tracker at Savannah [1] and get reviewed and
commented there, if maintainers or users are interested in your function. If
you don't get a fast response there, don't be disappointed, we are all
working on Octave as volunteers.

The `mapminmax` seems to be part of Matlab's Neural Network toolbox [2]. We
have an ongoing GSoC project about that package [3]. Maybe you can get in
contact with Enrico or his mentors Francesco or Ankit about your function?

Best,
Kai


[1]: https://savannah.gnu.org/patch/?group=octave
[2]: https://www.mathworks.com/help/nnet/ref/mapminmax.html
[3]: https://gsocnnet.blogspot.de/



--
View this message in context: 
http://octave.1599824.n4.nabble.com/Contribution-of-mapminmax-equivalent-function-implementation-tp4684239p4684244.html
Sent from the Octave - General mailing list archive at Nabble.com.