Fits Neural Networks to Learn About Backpropagation
Computes accuracy
Computes a gradient
Computes a gradient
Computes output
Computes output
Computes confusion matrix
Creates random weights
Creates random weights
Cross entropy
Finds best threshold
One step in backpropagation
One step in backpropagation
Checks for correct input
Checks for correct input
Checks for correct input
Checks for correct input
Checks for correct input
Checks for correct input
Checks for correct input
Checks for correct input
Checks for correct input
Checks for correct input
Differential of logistic function
Logistic function
Computes prediction
Computes prediction
Computes squared error
Sums up cross entropy
Sums up squared error
Fit neural networks with up to 2 hidden layers and one output neuron
Fits the neural network
Transforms prediction
Weights objects
Weights2 objects
Can fit neural networks with up to two hidden layer and two different error functions. Also able to handle a weight decay. But just able to compute one output neuron and very slow.