Answered
Getting best fitting model when using trainlm
Early stopping ONLY depends on the 15% validation subset performance. NOT on the 70% training or 15% testing performance. Inc...

7 years ago | 0

| accepted

Answered
How to solve the multiple regression problem with the help of neural networks in Matlab?
I interpret multiple regression to mean a linear regression involving multiple inputs and outputs. This can be solved by matrix ...

7 years ago | 0

Answered
How to create a target matrix for neural network?
The classic classifier target matrix for c classes contain columns of the unit matrix eye(c). The corresponding output layer ...

7 years ago | 0

Answered
Getting while using NARXnet. I have Qsimulated and Qobserved data. I can find error (Qobserved-Qsimulated). I am new to narnet. I have to predict future 5 values of forcasting. Data is in cell format and not in Times series. Please help.
You don't understand net = narxnet(1:2,1:2,10) ??? 1. READ THE DOCUMENTATION AND REPRODUCE THE EXAMPLES IN BOTH doc nar...

7 years ago | 0

Answered
Getting while using NARXnet. I have Qsimulated and Qobserved data. I can find error (Qobserved-Qsimulated). I am new to narnet. I have to predict future 5 values of forcasting. Data is in cell format and not in Times series. Please help.
Use the command whos to make sure all variables have the correct dimensions I have examples in BOTH comp.soft-sys.mat...

7 years ago | 0

Answered
How to continue training my neural network
In order to successfully continue training with new data, 1. Either a. The new data has similar summary statistics as...

7 years ago | 0

Answered
How to change SNR value of noise signal?
Multiply the noise samples by a constant before adding to the signal. *Thank you for formally accepting my answer* Greg

7 years ago | 0

Answered
How can we recover the network state at iteration T
In the distant past I'm pretty sure that I have checked, by using the error plot, that it is done automatically. *Thank you f...

7 years ago | 0

Answered
Avoiding Overfitting by Averaging Multiple Neural Network
The best way to mitigate overtraining an overfit net is MINIMIZE THE NUMBER OF HIDDEN NODES SUBJECT TO A MAXIMUM ALLOWED ERRO...

7 years ago | 1

Answered
How to train a time series dataset ?
Consider current as a function of time. The rest should be straightforward using FITNET. HOWEVER, YOU HAVE USED THE TERM T...

7 years ago | 0

Answered
i have set of data consist of 6 input variables and one output,how can i find formula between inputs and output?
A single hidden layer neural net solution can always be obtained in the form y = B2 + LW * tanh( B1 + IW * x ) However ...

7 years ago | 0

Answered
Performance of Feed Forward Neural Network
% 1. ALWAYS START CLEAN close all, clear all, clc % 2. ALWAYS INITIALIZE THE RNG SO THAT RESULTS CAN BE DUPLICATED !!! ...

7 years ago | 0

Question

MORE HELPFUL THAN CODE IN HELP/DOC/TYPE NEWFF?
% So many users are using NEWFF that I felt the following % would be useful (especially since it is also relevant % for...

7 years ago | 1 answer | 0

1

answer

Question

WHY HASN'T THE PERSISTENT ERROR IN HELP/DOC/TYPE NEWFF BEEN CORRECTED???
Although NEWFF is obsolete, it is still being used by many. Therefore, why hasn't the following coding ERROR in the docume...

7 years ago | 1 answer | 0

1

answer

Answered
During NN training how can restrict certain ouputs from going negative?
1. It really doesn't matter what values occur during training as long as the final values are correct. 2. Scale the outputs t...

7 years ago | 0

Answered
Using saved neural network on different size input
The previous answer is not correct. The topic of removing inputs that are correlated with other inputs is not trivial. Unf...

7 years ago | 2

| accepted

Answered
help me ? How many layers?
1. ONE HIDDEN LAYER IS ALWAYS SUFFICIENT! 2. NEWFF is obsolete, Do you have FITNET? help fitnet doc fitnet 2. If yo...

7 years ago | 0

Answered
How alignment of data is treated by the neural network algorithms for one day ahead prediction
I don't understand your problem. What fraction of the target variance did you want to achieve? mse(error)/mean(var(targe...

7 years ago | 0

Answered
Problem in data entry in neural network nntool
Do not use a NN for simple multiplication. If you are looking for an introductory problem see the examples at help fitnet...

7 years ago | 0

Answered
Patternnet for multi-dimensional classification
The training target should have nonnegative entries that sum to 1 and can be interpreted as prior probabilities. 0 and 1 are...

7 years ago | 0

| accepted

Answered
i have problem with ffnn on matlab
close all, clear, clc % N No. of input and output-target vectors % I Input dimension % O Output dimension %...

7 years ago | 0

Answered
Neural network coder decoder
I think I understand your problem. I have used the following technique 1. target = input 2. Single hidden layer 3. Use FIT...

7 years ago | 0

Answered
NARX/time series network for classification
Do not use Narx. Determine how many points are necessary for a prediction. Then form input vectors using that dimensionality...

7 years ago | 0

Answered
Computing Cross Entropy and the derivative of Softmax
Search both comp.soft-sys.matlab and ANSWERS for greg crossentropy Hope this helps. *Thank you for f...

7 years ago | 0

Answered
Neural Network non mutually exclusive cross entropy loss function
The target matrix columns should contain the NON-NEGATIVE UNIT SUM APRIORI CLASS PROBABILITIES !!! See any text r...

7 years ago | 0

| accepted

Answered
What threshold is plotconfusion applying?
% The example in the "help PLOTCONFUSION" dcumentation doesn't help because there are no errors with the simpleclass_dataset! Th...

7 years ago | 1

| accepted

Answered
[Neural network] How does neural network calculate output from net.IW, net.LW, net.b ?
You did not consider net.input.processFcns and net.output.processFcns Search in ANSWERS and comp.soft-sys.m...

7 years ago | 0

Answered
How to determine input delay in a Neural Network (NARX) ?
I have thoroughly discussed this in many COMP.SOFT-SYS.MATLAB and ANSWERS posts. Search in both using greg significant delay...

7 years ago | 0

Answered
In regression, do i need uniform distribution of output?
>> I understand why i need normalization but i don't understand unform distribution of output(=response) works weel without need...

7 years ago | 0

| accepted

Answered
is this NN using Backpropagation ??
1. NEWFF is an OBSOLETE function. Check the HELP and DOC documentation help newff doc newff For example: >> he...

7 years ago | 0