Neural Networks with R

Testing the network

We finally have the network trained and ready for use. Now, we can use it to make our predictions. Remember, we've set aside 30 percent of available data and then use them to test the network. It's time to use it.

predict_net_test <- compute(net,test_data[,1:9])

To predict data, we have used the compute function, which computes the outputs of all neurons for specific arbitrary covariate vectors, given a trained neural network. Let's look at the results by printing the first ten lines:

> head(predict_net_test$net.result,n=10)                 [,1]                       [,2]1  0.9999999935589190 0.0000000035872535107208482  0.0000011083596034 0.9999993767645581899117254  0.9792070465712006 0.0171647096645310796858565 0.9999999746453074 0.000000021909385204003642 ...