From: Edaj Juarez on
Good day

I have a backpropagation neural network for voice recognition, and sometimes for the number of the neurons the result is: "Validation stop" or "Performence met" and how to know when the NN is in that state, is important because i want to know which combination is ok and with that built fold validation.

Thanks

David
From: Greg Heath on
On Oct 14, 1:15 pm, "Edaj Juarez" <e...(a)walla.com> wrote:
> Good day
>
> I have a backpropagationneuralnetwork for voice recognition, and sometimes for the  number of the neurons the result is: "Validation stop" or "Performence met" and how to know when the NN is in that state, is important because i want to know which combination is ok and with that built fold validation.

Why isn't it obvious from the training plot?

Hope this helps.

Greg
 | 
Pages: 1
Prev: What is .csf file
Next: 64bits and memory problem