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

Wrong learning process?

Expand Messages
  • abi_11
    I have a problem with a network I just created, it s the first one I try to create and I really thing it did not learnt right. The network is supposed to be
    Message 1 of 5 , Apr 1, 2004
    • 0 Attachment
      I have a problem with a network I just created, it's the first one I
      try to create and I really thing it did not learnt right. The
      network is supposed to be able to determine the price of a car based
      on several factors like engine type, max speed… Well, when it comes
      to 60 to 0 and 80 to 0, the network is considering that the longer
      time the more expensive the car has to be. Is there a way to tell
      the network that it has to be the opposite? If not, what can I do?

      Tahnk you!
    • walt
      an easy way is to subtract the time from a maximum so the lower time is more valuable... For example: 30-14= 16 30-10= 20 The faster car (10 seconds) has a
      Message 2 of 5 , Apr 1, 2004
      • 0 Attachment
        an easy way is to subtract the time from a maximum so the lower time is more
        valuable...
        For example:

        30-14= 16
        30-10= 20

        The faster car (10 seconds) has a higher score...

        Walt
        -----Original Message-----
        From: abi_11 [mailto:abi_11@...]
        Sent: Thursday, April 01, 2004 9:01 AM
        To: easynn@yahoogroups.com
        Subject: [easynn] Wrong learning process?


        I have a problem with a network I just created, it's the first one I
        try to create and I really thing it did not learnt right. The
        network is supposed to be able to determine the price of a car based
        on several factors like engine type, max speed… Well, when it comes
        to 60 to 0 and 80 to 0, the network is considering that the longer
        time the more expensive the car has to be. Is there a way to tell
        the network that it has to be the opposite? If not, what can I do?

        Tahnk you!



        ----------------------------------------------------------------------------
        --
        Yahoo! Groups Links

        a.. To visit your group on the web, go to:
        http://groups.yahoo.com/group/easynn/

        b.. To unsubscribe from this group, send an email to:
        easynn-unsubscribe@yahoogroups.com

        c.. Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.



        [Non-text portions of this message have been removed]
      • walt
        Even easier; multiply time by -1 Walt ... From: abi_11 [mailto:abi_11@yahoo.com] Sent: Thursday, April 01, 2004 9:01 AM To: easynn@yahoogroups.com Subject:
        Message 3 of 5 , Apr 1, 2004
        • 0 Attachment
          Even easier; multiply time by -1

          Walt
          -----Original Message-----
          From: abi_11 [mailto:abi_11@...]
          Sent: Thursday, April 01, 2004 9:01 AM
          To: easynn@yahoogroups.com
          Subject: [easynn] Wrong learning process?


          I have a problem with a network I just created, it's the first one I
          try to create and I really thing it did not learnt right. The
          network is supposed to be able to determine the price of a car based
          on several factors like engine type, max speed… Well, when it comes
          to 60 to 0 and 80 to 0, the network is considering that the longer
          time the more expensive the car has to be. Is there a way to tell
          the network that it has to be the opposite? If not, what can I do?

          Tahnk you!



          ----------------------------------------------------------------------------
          --
          Yahoo! Groups Links

          a.. To visit your group on the web, go to:
          http://groups.yahoo.com/group/easynn/

          b.. To unsubscribe from this group, send an email to:
          easynn-unsubscribe@yahoogroups.com

          c.. Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.



          [Non-text portions of this message have been removed]
        • steve@easynn.com
          ... That made me realize that right-click functions like *-1 would not work! The test before execute will reject such functions. I ll put it on the bug list
          Message 4 of 5 , Apr 1, 2004
          • 0 Attachment
            On Thu, 01 Apr 2004 09:15:10 -0500, Walt wrote:

            >Even easier; multiply time by -1
            >

            That made me realize that right-click functions like *-1 would not
            work! The test before execute will reject such functions.

            I'll put it on the bug list but for now enter the value -1 in any
            unused grid cell. Right click the cell and "Set Function..." to *v
            Move to the column that needs to be changed, right click and "Execute
            on Input/Output Column".


            Steve
            --
            EasyNN-plus. The easy way to build neural networks.
            Build networks from numeric, text and image files.
            http://www.easynn.com
          • abi_11
            Awesome! I did notice what you said about the formulas not working… but I did the last thing you said and it worked. I did not really test the network yet,
            Message 5 of 5 , Apr 5, 2004
            • 0 Attachment
              Awesome! I did notice what you said about the formulas not working…
              but I did the last thing you said and it worked. I did not really
              test the network yet, but at least I got the number changed to
              minus… I'll see if it works!
              I'll let you know any further problems.

              Thank you!




              --- In easynn@yahoogroups.com, steve@e... wrote:
              > On Thu, 01 Apr 2004 09:15:10 -0500, Walt wrote:
              >
              > >Even easier; multiply time by -1
              > >
              >
              > That made me realize that right-click functions like *-1 would not
              > work! The test before execute will reject such functions.
              >
              > I'll put it on the bug list but for now enter the value -1 in any
              > unused grid cell. Right click the cell and "Set Function..." to *v
              > Move to the column that needs to be changed, right click
              and "Execute
              > on Input/Output Column".
              >
              >
              > Steve
              > --
              > EasyNN-plus. The easy way to build neural networks.
              > Build networks from numeric, text and image files.
              > http://www.easynn.com
            Your message has been successfully submitted and would be delivered to recipients shortly.