Databases
StackOverflow2013
Postsdbo
PONeuralNetwork back propagation question
 

Note that there are some explanatory texts on larger screens.

plurals
  1. PONeuralNetwork back propagation question
    primarykey
    data
    text
    <p>After reading a lot of other peoples neural network code I am convinced something is not right with my code. It works and I can train a network its just that in order to train the next perceptron in the hidden layer I must train the last one, shouldn't I be able to train all the units in a hidden layer in parallel?</p> <p>Here is the code it calculates the error of the hidden layer:</p> <pre><code> for(int i=n-&gt;numOfPerceptronLayers-2;i&gt;=1;i--) { // for all hidden layers float sum = 0.0; // &lt;- This here is the problem for(int j=0;j&lt;n-&gt;perceptronLayers[i].numOfPerceptrons;j++) { // For all the units in the current hidden layer for(int k=0;k&lt;n-&gt;perceptronLayers[i].perceptrons[j].numOfConnections;k++) { // Loop through the current units connections to the previous layer (output layer) sum += n-&gt;perceptronLayers[i+1].perceptrons[k].error * n-&gt;perceptronLayers[i+1].perceptrons[k].weights[j]; } n-&gt;perceptronLayers[i].perceptrons[j].error = n-&gt;perceptronLayers[i].perceptrons[j].output * (1.0 - n-&gt;perceptronLayers[i].perceptrons[j].output) * sum; } } </code></pre> <p>It should be like this (but this doesn't work):</p> <pre><code>for(int i=n-&gt;numOfPerceptronLayers-2;i&gt;=1;i--) { // for all hidden layers for(int j=0;j&lt;n-&gt;perceptronLayers[i].numOfPerceptrons;j++) { // For all the units in the current hidden layer float sum = 0.0; for(int k=0;k&lt;n-&gt;perceptronLayers[i].perceptrons[j].numOfConnections;k++) { // Loop through the current units connections to the previous layer (output layer) sum += n-&gt;perceptronLayers[i+1].perceptrons[k].error * n-&gt;perceptronLayers[i+1].perceptrons[k].weights[j]; } n-&gt;perceptronLayers[i].perceptrons[j].error = n-&gt;perceptronLayers[i].perceptrons[j].output * (1.0 - n-&gt;perceptronLayers[i].perceptrons[j].output) * sum; } } </code></pre> <p>Why is it that the sum variable must be declared for the entire layer rather than a single perceptron?</p>
    singulars
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. PTQuestion
    1. This table or related slice is empty.
    1. USA Person
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. PO
      singulars
      1. PTAnswer
    2. PO
      singulars
      1. PTAnswer
    1. VO
      singulars
      1. PONeuralNetwork back propagation question
      1. This table or related slice is empty.
      1. VTUpMod
    1. COCould you be more specific on what "it doesn't work" means. And could you add the exact mathematical formula you are trying to encode, because I have the impression something might get lost in translation here.
      singulars
      1. PONeuralNetwork back propagation question
      1. This table or related slice is empty.
    2. COThe mathematical formula is the back propagation learning, I am trying to calculate the errors of the perceptrons in the hidden layer. It doesn't work means that the training process does not work the network never learns what it is suppose to in the second piece of code.
      singulars
      1. PONeuralNetwork back propagation question
      1. USA Person
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload