Sensitivity Analysis of Neural Networks
Activation function of neuron
Define function to create a 'diagonal' array or get the diagonal of an...
Find Critical Value
NeuralSens: Sensitivity Analysis of Neural Networks
Sensitivity alpha-curve associated to MLP function
Sensitivity alpha-curve associated to MLP function of an input variabl...
Change significance of boot SensMLP Class
Sensitivity analysis plot over time of the data
Plot sensitivities of a neural network model
Plot sensitivities of a neural network model
Second derivative of activation function of neuron
Third derivative of activation function of neuron
Derivative of activation function of neuron
Define function to change the diagonal of array
Define function to create a 'diagonal' array or get the diagonal of an...
Define function to change the diagonal of array
Second derivatives 3D scatter or surface plot against input values
Feature sensitivity plot
Sensitivity of MLP models
Constructor of the HessMLP Class
Convert a HessMLP to a SensMLP object
Check if object is of class HessMLP
Check if object is of class SensMLP
k-StepM Algorithm for Hypothesis Testing
Plot method for the HessMLP Class
Plot method for the SensMLP Class
Neural network structure sensitivity plot
Print method for the HessMLP Class
Print method for the SensMLP Class
Print method of the summary HessMLP Class
Print method of the summary SensMLP Class
Sensitivity of MLP models
Sensitivity scatter plot against input values
Feature sensitivity plot
Plot sensitivities of a neural network model
Plot sensitivities of a neural network model
Constructor of the SensMLP Class
Sensitivity analysis plot over time of the data
Summary Method for the HessMLP Class
Summary Method for the SensMLP Class
Analysis functions to quantify inputs importance in neural network models. Functions are available for calculating and plotting the inputs importance and obtaining the activation function of each neuron layer and its derivatives. The importance of a given input is defined as the distribution of the derivatives of the output with respect to that input in each training data point <doi:10.18637/jss.v102.i07>.