Respondida
Neural net visualisation
The simplest solutions use integer weights, two hidden nodes with sign activation functions, purelin output activation functions...

casi 15 años hace | 0

| aceptada

Respondida
Data classification: Learning vector Quantization or two-layer feed-forward network???
LVQ is not self organizing. It is created using supervised learning. The FFMLP has a different topology and has been proven to b...

casi 15 años hace | 0

| aceptada

Respondida
More than one hidden layer
Dimitris used the older version of newff net = newff(minmax(p),[H1 H2 O],{'tansig' 'tansig' 'tansig'}); for a net with I-H...

casi 15 años hace | 0

| aceptada

Respondida
features selection problem..
What you are asking is not, in general, feasible. How many reference values are used for each decision? Assume there are ...

casi 15 años hace | 0

| aceptada

Respondida
Multiple output in TDNN (Time Delay Neural Network)
In the call to preparenets the input and output dimensions are automatically determined. Hope this helps. Greg

casi 15 años hace | 0

Respondida
problen in neural network
Please read the documentation and corresponding examples before asking elementary questions. Thank you. Greg P.S. If ...

casi 15 años hace | 0

Respondida
Weights during network training
You will have to loop over an inner one-epoch loop. Hope this helps. Greg

casi 15 años hace | 0

Respondida
plott the performance of the Neural Network
The training history is contained in the output structure tr. Use this to create your own plot. Hope this helps. Greg

casi 15 años hace | 0

Respondida
Value of Test error for nn
The output structure tr should contain that info ... [net tr ] = train(net,...); Hope this helps. Greg

casi 15 años hace | 0

Respondida
Neural network performance function, weighted sse, and false alarms
If you are designing a classifier with c classes, use training targets that are columns of eye(c). The input is assigned to the ...

casi 15 años hace | 0

| aceptada

Respondida
How to "retrain" Neural Networks?
Everytime you call newfit you create a new network with random initial weights. Therefore, try something like net = newfit...

casi 15 años hace | 0

Respondida
supervised training of SOM in MATLAB
The "SO" in SOM means "Self-Organizing" and refers to using the Kohonen algorithm for UNSUPERVISED clustering. Do not use the ac...

casi 15 años hace | 0

| aceptada

Respondida
how to set the number of hidden neurons in the neural network toolbox
This appears to be a classification problem; NOT a clustering problem. 1. 200 data points define a 199 dimensional space....

casi 15 años hace | 0

| aceptada

Respondida
Help capturing diagrams (printing) using Neural Network Toolbox
You have the option of modifying the source code or creating your own plots using tr. Hope this helps. Greg

casi 15 años hace | 0

| aceptada

Respondida
How to achieve good results in MLP
for i = 1: 6 T(i,1+9*(i-1):1+9*i) = ones(1,9); % vectorized version end You replaced Neq >> Nw with Neq = ...

casi 15 años hace | 0

| aceptada

Respondida
NN transfer function (tansig and purelin)
Several posters, including myself have been unable to reproduce outputs using the weights stored with the network. Hope th...

casi 15 años hace | 1

Respondida
How to use new learning algorithm for training ANN other than predefined algorithms (like FFBP)
Yes. Yes. Hope this helps. Greg

casi 15 años hace | 0

Respondida
problem of using patternnet in Neural network toolbox
use the pattern recognition network (patternnet) to train a set of data, whose dimension is 7100 and there have 2000 data poin...

casi 15 años hace | 0

| aceptada

Respondida
Help using MSE or MAE
mae is the mean of the absolute values of the errors. If you use mae for learning, then tabulate mae. mse is the mean of...

casi 15 años hace | 0

| aceptada

Respondida
Sample Neural network code
I have posted many examles of code. Search comp.soft-sys.MATLAB using heath close clear newff Hope this helps.

casi 15 años hace | 0

Respondida
Neural Networks-- train function error Indexing cannot yield multiple results.
Cannot use train as a filename or variable. Hope this helps. Greg

casi 15 años hace | 0

| aceptada

Respondida
How to get answers for my unanswered questions?
Try the newsgroup. Hope this helps. Greg

casi 15 años hace | 0

Respondida
data filtering for clustering
No. You are designing a classifier to classify unknown data. If each class was processed individually for training, you w...

casi 15 años hace | 0

| aceptada

Respondida
Neural net visualisation
You can print the weights. However, it is generally impossible to understand the weights of a regression MLP for a mediu...

casi 15 años hace | 0

Respondida
neural network toolbox
I answered this in the Newsgroup: 3 min 26 sec. Greg

casi 15 años hace | 0

| aceptada

Respondida
Testing Neural Network on new data
Here is my code: >[Pr,Pc] = size(PP);% Pr=486, Pc=1200 Pr is probably much larger than necessary. Try input variable subse...

casi 15 años hace | 1

Respondida
PCA matrix data compression help
For classification, choose the dimensions in the direction of greatest class separation. This is not guaranteed using PCA ...

casi 15 años hace | 2

| aceptada

Respondida
obtaining accuracy using neural netowrk
This appears to be a list of inputs. What outputs do you want? To design a NN you need a large diverse sample of input/...

casi 15 años hace | 0

| aceptada

Respondida
Picking an FFT length
Why not just overlay the times vs size plot for X = fft(randn(N,M)) when M = 1e4 N1 = 2.^(1:10) % blue and N2 = floor(...

casi 15 años hace | 0

| aceptada

Respondida
retrain neurons
You could try [net2 tr Y E] = train(net1,p2,t2); % Train for a new set % However, performance of net2 on p1 may be unsatis...

casi 15 años hace | 1

| aceptada