SSLR (version 0.9.3.3)

democratic: General Interface for Democratic model

Description

Democratic Co-Learning is a semi-supervised learning algorithm with a co-training style. This algorithm trains N classifiers with different learning schemes defined in list gen.learners. During the iterative process, the multiple classifiers with different inductive biases label data for each other.

Usage

democratic(learners, schemes = NULL)

Arguments

learners

List of models from parsnip package for training a supervised base classifier using a set of instances. This model need to have probability predictions

schemes

List of schemes (col x names in each learner). Default is null, it means that learner uses all x columns

Value

(When model fit) A list object of class "democratic" containing:

W

A vector with the confidence-weighted vote assigned to each classifier.

model

A list with the final N base classifiers trained using the enlarged labeled set.

model.index

List of N vectors of indexes related to the training instances used per each classifier. These indexes are relative to the y argument.

instances.index

The indexes of all training instances used to train the N models. These indexes include the initial labeled instances and the newly labeled instances. These indexes are relative to the y argument.

model.index.map

List of three vectors with the same information in model.index but the indexes are relative to instances.index vector.

classes

The levels of y factor.

preds

The functions provided in the preds argument.

preds.pars

The set of lists provided in the preds.pars argument.

x.inst

The value provided in the x.inst argument.

Details

This method trains an ensemble of diverse classifiers. To promote the initial diversity the classifiers must represent different learning schemes. When x.inst is FALSE all learners defined must be able to learn a classifier from the precomputed matrix in x. The iteration process of the algorithm ends when no changes occurs in any model during a complete iteration. The generation of the final hypothesis is produced via a weigthed majority voting.

Examples

# NOT RUN {
library(tidyverse)
library(tidymodels)
library(caret)
library(SSLR)

data(wine)

set.seed(1)
train.index <- createDataPartition(wine$Wine, p = .7, list = FALSE)
train <- wine[ train.index,]
test  <- wine[-train.index,]

cls <- which(colnames(wine) == "Wine")

#% LABELED
labeled.index <- createDataPartition(wine$Wine, p = .2, list = FALSE)
train[-labeled.index,cls] <- NA

#We need a model with probability predictions from parsnip
#https://tidymodels.github.io/parsnip/articles/articles/Models.html
#It should be with mode = classification


rf <-  rand_forest(trees = 100, mode = "classification") %>%
  set_engine("randomForest")


bt <-  boost_tree(trees = 100, mode = "classification") %>%
  set_engine("C5.0")


m <- democratic(learners = list(rf,bt)) %>% fit(Wine ~ ., data = train)

#' \donttest{
#Accuracy
predict(m,test) %>%
  bind_cols(test) %>%
  metrics(truth = "Wine", estimate = .pred_class)


#With schemes
set.seed(1)
m <- democratic(learners = list(rf,bt),
                schemes = list(c("Malic.Acid","Ash"), c("Magnesium","Proline")) ) %>%
  fit(Wine ~ ., data = train)


#Accuracy
predict(m,test) %>%
  bind_cols(test) %>%
  metrics(truth = "Wine", estimate = .pred_class)

#'}
# }