Deep in the control logic of EasyNN-plus is a diagnostic facility that adds a delay to each cycle so that learning can be slowed down. It was first included so...
... The number of milliseconds to delay each cycle is set in the Controls. It must also be enabled with a checkbox so it's difficult to put on unintentionally....
As the culprit responsible for this suggestion, I guess I should respond. What I am interested in is showing a class of beginners (1st year undergrads) how a...
I had an hour to spare so slow learning is implemented and now available. The 3.0i release has a cycle delay and checkbox in the Controls dialog. To switch on...
Cool! Look forward to trying it out with my students Steve ... http://www.easynn.com ... Ink Cartridges or Refill Kits for your HP, Epson, Canon or Lexmark...
sir, i wish to appreciate your effort on making me be a menber of thes group. akin. ... Do you Yahoo!? The New Yahoo! Shopping - with improved product search ...
Today I received a network file produced by EasyNN that could not be opened by EasyNN-plus. The problem is that EasyNN-plus checks all the rows to see if the ...
While investigating the problem with grids greater than 16,384 rows a number of other performance issues have been identified. The most serious one is removing...
Steve, I have a method to remove dupilcates wich takes very little time: 1. create a csv file of the input data 2. import it into "microsoft access" 3. create...
... My message could have been confusing. By duplicates I meant the ones produced by Edit > Duplicate Examples Rows... They use the row name prefix D: and...
A recent support message says "When the network learning is started sometimes it says "Starting" but never does anything. It won't stop starting even if I kill...
Could someone suggest a way to determine how confident the network is that its output/prediction will be within a certain range of the true outcome. Thank You...
... output to ... Well, if I give it a string of random numbers, after it trains it's very familiar with what it's already seen so . . . doesn't comparing the...
It depends on the structure of the net and what your training stop criteria was. In my case, this technique does help me to determin if one network ...
... Networks can learn from random data but will fail with validating data. See "Random Date.tvq". Have a look at the Predictions view. Steve -- EasyNN-plus....
I am not sure if I am on or off the initial track here but suppose I ask "you" to learn from a series of random nubers and "predict" the continuation of the...
... The random tests I use are groups of five values rather than series. That is a five input and five output network with eight hidden nodes. To me, a series...
Steve, Walt, Ken Thanks for the input, Maybe random wasn't the best choice of words on my part. I'll let you in on how I thought I might apply such a tool. ...
... It would be possible to produce a network that could learn a series of moves for one person if they always play in predictable way. Most people would try...
... of ... move. ... I'm doing more of a time-series kind of thing than finding a rotating pattern: I use the following inputs: -opponent rock move -opponent...
For time series predictions one would need an Input/Output type variable which is lacking in EasyNN Plus. For example, if you have data on four variables...
... I think you will find Copy Input/Output Columns... will do most of the work for you. Copy the inputs to outputs with an offset of -1. Exclude the first and...
Hi Steve, Unless I've misunderstood you, I don't think that Copy "Input/Output Columns" can take care of this situation because your grid will grow ...
... That's true if a pair of inputs and outputs is produced for every example row in the series. I was suggesting that only one pair was created so row n...
Having done time series prediction manually in easyNN, I would certainly find such a feature useful. There was an example about a 12 variable time series over...
In fact this is a fantastic feature to have in EasyNN-Plus. Ideally the tool should also be able to see the effects of all variables (or determine a subset of...
Malekz et al, In time series prediction, if you used the output to feed the input, it is very doubtful that the result would be as good as using the inputs to ...
Hi Walt, What you say maybe valid for your stock price application, but we still need an Input/Output variable for a little bit more complicated applications....
Having problems with message search?
Fill out this form to ensure your group is one of the first to be migrated to the new message search system.
