[vorbis-dev] questions re residuevqtrain

Gregory Maxwell greg at linuxpower.cx
Fri Dec 22 05:02:03 PST 2000

On Fri, Dec 22, 2000 at 04:30:01AM -0600, Mike Coleman wrote:
> > Yeah, things were already bad at this point (all the dupes).  In this case,
> > the data file is probably way too small to train (not enough short blocks to
> > produce a set).
> What does "produce a set" mean?  A set of what?
> (The residue file had ~3700 training points, I think, to train 256 entries.)

The vector trainer tools will probably never converge unless you have at
least 16 entries per cell (I supposes it depends on the standard deviation
of the input, but I've never had good results with <16  entries per cell)

--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to 'vorbis-dev-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.

More information about the Vorbis-dev mailing list