Table 1 Classification algorithms and their parameterization
From: Towards a supervised classification of neocortical interneuron morphologies
Classifier | Abbreviation | R Package | Prespecified Parameters |
|---|---|---|---|
Classification and regression trees | CART | rpart [71] | \(|\mathcal {D}^{a}|\) = 10, \(|\mathcal {D}^{l}| = 5\) |
k nearest neighbors | kNN | kknn [72] | k=5,p=2 unweighted |
Linear discriminant analysis | LDA | MASS [73] | |
Gaussian naive Bayes | NB | e1071 [74] | |
Random forest | RF | randomForest [75] | \(T = 2000, m = \sqrt {n}\) |
Lasso regularized logistic regression | RMLR | glmnet [76] | λ=0.01 |
Support vector machine | SVM | RBF: \(\gamma = \frac {1}{n}, C = 1\) | |
Single-layer neural network | NNET | neuralnet [78] | h=5 |
AdaBoost | ADA | gbm [79] | T=3000d=1s=0.001 |