Press escape to exit fullscreen

{{sketch.instructions}}

CC {{sketch.licenseObject.short}}

Archived Sketch

This sketch is created with an older version of Processing,
and doesn't work on browsers anymore.

View Source Code

Capture Screenshot

SOM_3d

{{$t('general.by')}}
Miriam
Dall'Igna
SOM are a manner of representing multidimensional data in much lower dimensional spaces through data compression by the reduction of the dimensionality of vectors known as vector quantisation. These networks store information in such a way that any topological relationships within the data set are maintained. The interesting aspect of these networks is that they learn how to classify data without supervision. Therefore as opposed to other ANN there is no target output vector. References: http://www.eicstes.org/EI http://www.ai-junkie.com/ann/som/som1.html http://www.jjguy.com/som/
We recovered an unsaved version of this sketch. Please review your changes below.

As a Plus+ Member feature, this source code is hidden by the owner.

  • {{co.title}}
    •
    {{$t('sketch.mode-pjs')}} {{$t('general.learnMore')}}
    Select mode or a template
    {{liveCodeLink}}
    • {{l.url.substr(l.url.lastIndexOf('/') + 1)}}
    {{$t('sketch.layoutDescription')}}

    {{k.category}}
    {{k.label}}
    {{k.description}}

    {{$t('sketch.seeMoreShortcuts')}}

    Versions are only kept for 7 days.
    Join Plus+ to keep versions indefinitely!

    {{$t('general.joinPlus')}}

    {{$t('sketch.versionsSummarized')}}

    {{ $t('general.joinPlus') }}