[R] Median of streaming data

Mohan Radhakrishnan radhakrishnan.mohan at gmail.com
Tue Sep 30 09:57:08 CEST 2014

       I came across this project(http://jwijffels.github.io/RMOA/) which
seems to be directly addressing streaming prediction.

>From the manual :

A data stream environment has different requirements from the traditional
setting. The most significant are the following:
Requirement 1 Process an example at a time, and inspect it only once (at
Requirement 2 Use a limited amount of memory
Requirement 3 Work in a limited amount of time
Requirement 4 Be ready to predict at any time


On Sat, Sep 27, 2014 at 4:23 AM, Rolf Turner <r.turner at auckland.ac.nz>

> On 26/09/14 21:48, Martin Maechler wrote:
>> Rolf Turner <r.turner at auckland.ac.nz>
> <SNIP>
>>      > I have coded up the algorithm from the Cameron and Turner
>>      > paper.  Dunno if it gives exactly the same results as my
>>      > (Splus?) code from lo these many years ago (the code that
>>      > is lost in the mists of time), but it *seems* to work.
>> excellent, thank you, Rolf!
>>      > It is not designed to work with actual "streaming" data
>>      > --- I don't know how to do that.  It takes a complete data
>>      > vector as input.  Someone who knows about streaming data
>>      > should be able to adapt it pretty easily.  Said he, the
>>      > proverbial optimist.
>> I agree; that should not be hard.
>> One way is to replace   'y[ind]' by   'getY(ind)' everywhere in the code
>> and let 'getY' be an argument to rlas() provided by the user.
>>      > The function code and a help file are attached.  These
>>      > files have had their names changed to end in ".txt" so
>>      > that they will get through the mailing list processor
>>      > without being stripped.  With a bit of luck.
>> ;-)
>> It did work indeed.
>> I've added them to  'robustX' -- on R-forge,
>> including a plot() method and some little more flexibility.
>>    --> https://r-forge.r-project.org/R/?group_id=59
> <SNIP>
> Since I posted my previous email I have found some typos in the
> documentation and made some adjustments to the code.
> I also realized that the name "rlas" sounds a bit too much like a random
> number generator, according to R's conventions.  So I have decided that
> "lasr" would be a better name.
> I hope this change of horses in midstream doesn't mess things up too much
> for you.  If it does, of course feel free to stick with "rlas".
> I have attached revised *.R and *.Rd files.  Not sure that the *.Rd file
> will get through as-is.  Please let me know if it doesn't.
> cheers,
> Rolf
> --
> Rolf Turner
> Technical Editor ANZJS

	[[alternative HTML version deleted]]

More information about the R-help mailing list