Demonstration 1 Loading a File and Running a CSV Test Data Set - adam-nnl/Erudite GitHub Wiki

For this demonstration we will use an already trained neural network to predict NFL football games. This is part of the AFootball sub-project.

  1. Go into your Erudite-v1.1.0-Full_Suite folder and launch ‘Erudite-v1.1.0.jar‘

  2. After the program loads click the ‘Browse…‘ button in the ‘File Settings‘ section

  3. Navigate into the ‘AFootball‘ directory and then into the ‘ANNeML Files‘ sub-directory

  4. select/open ‘AFNNET_ANNeML_logsig_6H1Lv2-rand-TRAINED.xml‘

  5. Go to ‘Input Settings‘ section and click the ‘Import Batch‘ button

  6. Navigate into the ‘AFootball‘ directory and select/open ‘nfl_weeklystats_week7.csv‘ or download and use the NFL week 8 stats

  7. Click the ‘Run NNet‘ button, all of the data sets(games) will be processed through the neural network and the resulting outputs will print out. Outputs of approximately one(1.00-0.70) represents a win and outputs of approximately zero(0.30-0.00) represent a loss

Further info about predicting NFL games with the AFootball project can be found here or here

Continue to demo 2a