Loading ...
Sorry, an error occurred while loading the content.

EasyNN-plus V11.0a

Expand Messages
  • Stephen Wolstenholme
    EasyNN-plus V11.0a is available from today. The changes are as follows:- 1. Hidden node Cloning. This can help when there are problems finding a neural network
    Message 1 of 3 , Jun 7, 2009
    • 0 Attachment
      EasyNN-plus V11.0a is available from today.

      The changes are as follows:-

      1. Hidden node Cloning. This can help when there are problems finding
      a neural network configuration that will learn a complex dataset.
      Neural networks can be built with cloning enabled or cloning can be
      enabled while the network is learning. Hidden nodes are cloned during
      the learning process when the average error stops reducing. The hidden
      node with the greatest net input is frozen to determine if it is
      contributing to the learning process. If the node is contributing to
      the learning process it is cloned. Nodes that are not contributing are
      skipped and the node with the next greatest net input is tested. If no
      suitable node can be found a new node is created and cloned. The node
      that has been cloned is frozen again and its clone is immediately
      included in the learning process. The freeze level of the node that
      has been cloned is reduced until both the node and its clone are
      contributing to the learning process. The learning process does not
      need to restart when a node is cloned.

      2. Added "Set cloning" to New Network dialog to enable cloning on any
      network.

      3. Added "Cloning" toolbar button to enable hidden node cloning while
      the network is learning.

      4. Changed accelerator to have a much greater influence on the
      learning speed.

      5. Changed network view to show all weights regardless of their
      freeze level.

      6. Moved "Special Files" from the Tools menu to the File menu and
      renamed as Export.

      7. Improved navigation tracing.

      8. Made import csv/text function recognize either CR or LF as
      newline.

      9. Fixed a bug with Accelerator decay.

      10. Fixed a problem with remembering excluded columns on the column
      values graph.

      The full version has already been uploaded.

      The trial version will be available later today.

      Steve

      --
      Neural Planner Software Ltd www.NPSL1.com
    • P. Dickens
      Hi Steve, May I have the URL please. This sounds a really interesting development. I have just the application to test it on, which has been giving me some
      Message 2 of 3 , Jun 7, 2009
      • 0 Attachment
        Hi Steve,

        May I have the URL please.

        This sounds a really interesting development. I have just the
        application to test it on, which has been giving me some severe
        problems.

        Many thanks,

        Peter


        --------------------------------
        Stephen Wolstenholme wrote:

        > EasyNN-plus V11.0a is available from today.
        >
        > The changes are as follows:-
        >
        > 1. Hidden node Cloning. This can help when there are problems finding
        > a neural network configuration that will learn a complex dataset.
        > Neural networks can be built with cloning enabled or cloning can be
        > enabled while the network is learning. Hidden nodes are cloned during
        > the learning process when the average error stops reducing. The hidden
        >
        > node with the greatest net input is frozen to determine if it is
        > contributing to the learning process. If the node is contributing to
        > the learning process it is cloned. Nodes that are not contributing are
        >
        > skipped and the node with the next greatest net input is tested. If no
        >
        > suitable node can be found a new node is created and cloned. The node
        > that has been cloned is frozen again and its clone is immediately
        > included in the learning process. The freeze level of the node that
        > has been cloned is reduced until both the node and its clone are
        > contributing to the learning process. The learning process does not
        > need to restart when a node is cloned.
        >
        > 2. Added "Set cloning" to New Network dialog to enable cloning on any
        > network.
        >
        > 3. Added "Cloning" toolbar button to enable hidden node cloning while
        > the network is learning.
        >
        > 4. Changed accelerator to have a much greater influence on the
        > learning speed.
        >
        > 5. Changed network view to show all weights regardless of their
        > freeze level.
        >
        > 6. Moved "Special Files" from the Tools menu to the File menu and
        > renamed as Export.
        >
        > 7. Improved navigation tracing.
        >
        > 8. Made import csv/text function recognize either CR or LF as
        > newline.
        >
        > 9. Fixed a bug with Accelerator decay.
        >
        > 10. Fixed a problem with remembering excluded columns on the column
        > values graph.
        >
        > The full version has already been uploaded.
        >
        > The trial version will be available later today.
        >
        > Steve
        >
        > --
        > Neural Planner Software Ltd www.NPSL1.com
        >
        >


        [Non-text portions of this message have been removed]
      • Stephen Wolstenholme
        The trial version of EasyNN-plus v11.0a is now available. It can be downloaded from http://www.easynn.com/108491/ennsetup.exe Steve
        Message 3 of 3 , Jun 7, 2009
        • 0 Attachment
          The trial version of EasyNN-plus v11.0a is now available.

          It can be downloaded from http://www.easynn.com/108491/ennsetup.exe

          Steve
        Your message has been successfully submitted and would be delivered to recipients shortly.