PredictRBM: Predict RBM
In TimoMatzen/RBM: Package for fitting RBM and DBN models in R.

Description Usage Arguments Value Examples

Function for predicting on a test set with either a classification RBM (trained with the RBM function) or a stacked RBM with a classification RBM on top (trained with StackRBM).

1	PredictRBM(test, labels, model, layers = 2)

`test`	Is the test-data (matrix) on which the user wants to make predictions.
`labels`	Is a matrix with the corresponding labels for test-data.
`model`	Is the trained RBM or StackRBM model.
`layers`	Only needed with StackRBM to define the number of layers.

A list with a confusion matrix and the accuracy of the predictions.

# Load MNIST data
data(MNIST)

# First train supervised RBM or StackedRBM
mod <- RBM(MNIST$trainX, MNIST$trainY, n.iter = 1000, n.hidden = 1000)

# Then use PredictRBM to make predictions
PredictRBM(MNIST$testX, MNIST$testY, model = mod)