A | B | C | |
---|---|---|---|
1 | Filename | Demos featured in | Description/Notes |
2 | ml_binaryclass_bagging | demo_binaryclass_bagging | Binary classification based on the highest prediction among models to bootstrap samples. |
3 | ml_binaryclass_basis | demo_binaryclass_basis, demo_binaryclass_binaryclass_CV | Binary classification after a change of basis |
4 | ml_binaryclass_boosting | demo_binaryclass_boosting | Implements AdaBoost or LogitBoost algorithm to fit a binary classifier |
5 | ml_binaryclass_brokenStump | demo_binaryclass_exponential | Finds the best threshold across all features |
6 | ml_binaryclass_Cauchit | demo_binaryclass_GLM | Classifies data using generalized linear model with a Cauchit link |
7 | ml_binaryclass_CV | demo_binaryclass_CV | This computes the "best" hyper-parameter(s) using cross-validation for classification and regression problems |
8 | ml_binaryclass_exponential | demo_binaryclass_exponential | Fits a linear classifier by minimizing the exponential loss |
9 | ml_binaryclass_extreme | demo_binaryclass_GLM | Classifies data using generalized linear model with a extreme link |
10 | ml_binaryclass_HSVM | demo_binaryclass_SVM | Fits a linear classifier using Huberized SVM |
11 | ml_binaryclass_logistic | demo_binaryclass_bagging, demo_binaryclass_basis, demo_binaryclass_binaryclass_CV, demo_binaryclass_GLM | Fits a classifier using logistic regression with logistic output layer and logistic loss |
12 | ml_binaryclass_MLP | demo_binaryclass_NN | Binary classification using a multilayer perceptron with logistic loss |
13 | ml_binaryclass_multiclass | demo_binaryclass_alt | Using a multiclass classifier and treating the special case of binary labels |
14 | ml_binaryclass_perceptron | demo_binaryclass_NN | Fits a binary linear classifier by using the perceptron learning algorithm |
15 | ml_binaryclass_probit | demo_binaryclass_GLM | Fits a probit regression classifier by minimizing the negative likelihood |
16 | ml_binaryclass_randomForest | demo_binaryclass_decisions | A bagging of tree classifers, where the splits are chosen among random selection of features |
17 | ml_binaryclass_regression | demo_binaryclass_alt | Classification by using a regression model |
18 | ml_binaryclass_SSVM | demo_binaryclass_SVM | Fits a smooth support vector machine by minimizing squared hinge loss |
19 | ml_binaryclass_stump | demo_binaryclass_bagging, demo_binaryclass_boosting, demo_binaryclass_decisions | Finds the best threshold across all features |
20 | ml_binaryclass_SVM | demo_binaryclass_SVM | Fits a linear classifier by maximizing the margin using SVM |
21 | ml_binaryclass_tree | demo_binaryclass_decisions | Binary classification using a decision tree. The tree is constructed using the C4.5 algorithm, based on maximizing information gain |