Test Run - Autoencoders for Visualization Using CNTK RRS feed

  • General discussion

  • Graphing data that has just two dimensions is easy, but what if your data has multiple dimensions? James McCaffrey shows how you can create a CNTK neural network autoencoder to condense each item down to two dimensions, in order to enhance visualization of the data.

    Read this article in the December 2018 issue of MSDN Magazine

    Monday, December 3, 2018 7:53 PM