Neuron activation functions used for neural network nonlinearities.
Collaboration diagram for Neuron activation functions:
Classes | |
| struct | shark::TanhNeuron |
| Neuron which computes the hyperbolic tangenst with range [-1,1]. More... | |
| struct | shark::LogisticNeuron |
| Neuron which computes the Logistic (logistic) function with range [0,1]. More... | |
| struct | shark::FastSigmoidNeuron |
| Fast sigmoidal function, which does not need to compute an exponential function. More... | |
| struct | shark::LinearNeuron |
| Linear activation Neuron. More... | |
| struct | shark::RectifierNeuron |
| Rectifier Neuron f(x) = max(0,x) More... | |
| struct | shark::NormalizerNeuron< VectorType > |
| Normalizes the sum of inputs to one. More... | |
| struct | shark::SoftmaxNeuron< VectorType > |
| Computes the softmax activation function. More... | |