Demonstration 2a Loading a File and Training with a CSV File - adam-nnl/Erudite GitHub Wiki

For this demo we will train a simple backpropogation neural network with the XOR logic truth table.

  1. Go into your Erudite-v1.1.0-Full_Suite folder and launch β€˜Erudite-v1.1.0.jarβ€˜

  2. After the program loads click the β€˜Browseβ€¦β€˜ button in the β€˜File Settingsβ€˜ section

  3. Navigate into the β€˜ANNeML Examplesβ€˜ directory and select/open β€˜logicTest-alllogsig_ANNeML.xmlβ€˜

  4. Go to the β€˜Training Settingsβ€˜ section and check the box for β€˜Enable Trainingβ€˜

  5. Click the now-enabled button β€˜Import Training Setβ€˜

  6. Navigate to the β€˜Training Setsβ€˜ directory and select/open β€˜XOR_TS-logsig.csvβ€˜

  7. Select the option for β€˜Avg. error< 0.10β€˜

  8. Click the β€˜Train NNetβ€˜ button. The neural network with now go through as many training cycles as it needs to to β€œlearn” the training set.

  9. Once training has completed uncheck the β€˜Enable Trainingβ€˜ box to test your results with the next demonstration…


Continue to Demonstration 2b