io.github.mandar2812.dynaml.models.svm
Learn the parameters of the model.
Learn the parameters of the model.
Predict the value of the target variable given a point.
Predict the value of the target variable given a point.
Implements the changes in the model after application of a given kernel.
Implements the changes in the model after application of a given kernel.
It calculates
1) Eigen spectrum of the kernel
2) Calculates an approximation to the non linear feature map induced by the application of the kernel
A kernel object.
The number of prototypes to select in order to approximate the kernel matrix.
A Map which stores the current state of the system.
A Map which stores the current state of the system.
Calculates the energy of the configuration, in most global optimization algorithms we aim to find an approximate value of the hyper-parameters such that this function is minimized.
Calculates the energy of the configuration, in most global optimization algorithms we aim to find an approximate value of the hyper-parameters such that this function is minimized.
The value of the hyper-parameters in the configuration space
Optional parameters about configuration
Configuration Energy E(h)
The non linear feature mapping implicitly defined by the kernel applied, this is initialized to an identity map.
The non linear feature mapping implicitly defined by the kernel applied, this is initialized to an identity map.
The training data
The training data
Stores the names of the hyper-parameters
Stores the names of the hyper-parameters
Calculate an approximation to the subset of size M with the maximum entropy.
Calculate an approximation to the subset of size M with the maximum entropy.
Get the value of the parameters of the model.
Get the value of the parameters of the model.
This variable stores the indexes of the prototype points of the data set.
This variable stores the indexes of the prototype points of the data set.
Implementation of the Fixed Size Kernel based LS SVM
Fixed Size implies that the model chooses a subset of the original data to calculate a low rank approximation to the kernel matrix.
Feature Extraction is done in the primal space using the Nystrom approximation.