Help for package milr

Type:

Package

Title:

Multiple-Instance Logistic Regression with LASSO Penalty

Version:

0.3.1

Date:

2020-10-28

Description:

The multiple instance data set consists of many independent subjects (called bags) and each subject is composed of several components (called instances). The outcomes of such data set are binary or categorical responses, and, we can only observe the subject-level outcomes. For example, in manufacturing processes, a subject is labeled as "defective" if at least one of its own components is defective, and otherwise, is labeled as "non-defective". The 'milr' package focuses on the predictive model for the multiple instance data set with binary outcomes and performs the maximum likelihood estimation with the Expectation-Maximization algorithm under the framework of logistic regression. Moreover, the LASSO penalty is attached to the likelihood function for simultaneous parameter estimation and variable selection.

Author:

Ping-Yang Chen [aut, cre], ChingChuan Chen [aut], Chun-Hao Yang [aut], Sheng-Mao Chang [aut]

URL:

https://github.com/PingYangChen/milr

BugReports:

https://github.com/PingYangChen/milr/issues

Maintainer:

Ping-Yang Chen <pychen.ping@gmail.com>

Depends:

R (≥ 3.2.3)

Imports:

utils, pipeR (≥ 0.5), numDeriv, glmnet, Rcpp (≥ 0.12.0), RcppParallel

LinkingTo:

Rcpp, RcppArmadillo, RcppParallel

Suggests:

testthat, knitr, Hmisc, rmarkdown, data.table, ggplot2, plyr

LazyData:

yes

SystemRequirements:

GNU make

License:

MIT + file LICENSE

RoxygenNote:

7.1.1

VignetteBuilder:

knitr

NeedsCompilation:

yes

Packaged:

2020-10-28 09:52:41 UTC; rbchen

Repository:

CRAN

Date/Publication:

2020-10-31 07:30:02 UTC

The milr package: multiple-instance logistic regression with lasso penalty

Description

The multiple instance data set consists of many independent subjects (called bags) and each subject is composed of several components (called instances). The outcomes of such data set are binary or multinomial, and, we can only observe the subject-level outcomes. For example, in manufactory processes, a subject is labeled as "defective" if at least one of its own components is defective, and otherwise, is labeled as "non-defective". The milr package focuses on the predictive model for the multiple instance data set with binary outcomes and performs the maximum likelihood estimation with the Expectation-Maximization algorithm under the framework of logistic regression. Moreover, the LASSO penalty is attached to the likelihood function for simultaneous parameter estimation and variable selection.

References

Chen, R.-B., Cheng, K.-H., Chang, S.-M., Jeng, S.-L., Chen, P.-Y., Yang, C.-H., and Hsia, C.-C. (2016). Multiple-Instance Logistic Regression with LASSO Penalty. arXiv:1607.03615 [stat.ML].

DGP: data generation

Description

Generating the multiple-instance data set.

Usage

DGP(n, m, beta)

Arguments

n

an integer. The number of bags.

m

an integer or vector of length n. If m is an integer, each bag has the identical number of instances, m. If m is a vector, the ith bag has m[i] instances.

beta

a vector. The true regression coefficients.

Value

a list including (1) bag-level labels, Z, (2) the design matrix, X, and (3) bag ID of each instance, ID.

Examples

data1 <- DGP(50, 3, runif(10, -5, 5))
data2 <- DGP(50, sample(3:5, 50, TRUE), runif(10, -5, 5))

Fitted Response of milr Fits

Description

Fitted Response of milr Fits

Usage

## S3 method for class 'milr'
fitted(object, type = "bag", ...)

Arguments

object

A fitted obejct of class inheriting from "milr".

type

The type of fitted response required. Default is "bag", the fitted labels of bags. The "instance" option returns the fitted labels of instances.

...

further arguments passed to or from other methods.

Fitted Response of softmax Fits

Description

Fitted Response of softmax Fits

Usage

## S3 method for class 'softmax'
fitted(object, type = "bag", ...)

Arguments

object

A fitted obejct of class inheriting from "softmax".

type

The type of fitted response required. Default is "bag", the fitted labels of bags. The "instance" option returns the fitted labels of instances.

...

further arguments passed to or from other methods.

logit link function

Description

calculate the values of logit link

Usage

logit(X, beta)

Arguments

X

A matrix, the design matrix.

beta

A vector, the coefficients.

Value

An vector of the values of logit link.

Maximum likelihood estimation of multiple-instance logistic regression with LASSO penalty

Description

Please refer to milr-package.

Usage

milr(
  y,
  x,
  bag,
  lambda = 0,
  numLambda = 20L,
  lambdaCriterion = "BIC",
  nfold = 10L,
  maxit = 500L
)

Arguments

y

a vector. Bag-level binary labels.

x

the design matrix. The number of rows of x must be equal to the length of y.

bag

a vector, bag id.

lambda

the tuning parameter for LASSO-penalty. If lambda is a real value number, then the milr fits the model based on this lambda value. Second, if lambda is vector, then the optimal lambda value would be be chosen based on the optimality criterion, lambdaCriterion. Finally, if lambda = -1, then the optimal lambda value would be chosen automatically. The default is 0.

numLambda

An integer, the maximum length of LASSO-penalty. in atuo-tunning mode (lambda = -1). The default is 20.

lambdaCriterion

a string, the used optimality criterion for tuning the lambda value. It can be specified with lambdaCriterion = "BIC" or lambdaCriterion = "deviance".

nfold

an integer, the number of fold for cross-validation to choose the optimal lambda when lambdaCriterion = "deviance".

maxit

an integer, the maximum iteration for the EM algorithm. The default is 500.

Value

An object with S3 class "milr".

lambdaa vector of candidate lambda values.
cva vector of predictive deviance via nfold-fold cross validation when lambdaCriterion = "deviance".
deviancea vector of deviance of candidate model for each candidate lambda value.
BICa vector of BIC of candidate model for each candidate lambda value.
best_indexan integer, indicates the index of the best model among candidate lambda values.
best_modela list of the information for the best model including deviance (not cv deviance), BIC, chosen lambda, coefficients, fitted values, log-likelihood and variances of coefficients.

Examples

set.seed(100)
beta <- runif(5, -5, 5)
trainData <- DGP(40, 3, beta)
testData <- DGP(5, 3, beta)
# default (not use LASSO)
milr_result <- milr(trainData$Z, trainData$X, trainData$ID)
coef(milr_result)      # coefficients
fitted(milr_result)                    # fitted bag labels
fitted(milr_result, type = "instance") # fitted instance labels
summary(milr_result)   # summary milr
predict(milr_result, testData$X, testData$ID)                    # predicted bag labels
predict(milr_result, testData$X, testData$ID, type = "instance") # predicted instance labels

# use BIC to choose penalty (not run)
#milr_result <- milr(trainData$Z, trainData$X, trainData$ID,
#                    exp(seq(log(0.01), log(50), length = 30)))
#coef(milr_result)      # coefficients
#fitted(milr_result)                    # fitted bag labels
#fitted(milr_result, type = "instance") # fitted instance labels
#summary(milr_result)   # summary milr
#predict(milr_result, testData$X, testData$ID)                    # predicted bag labels
#predict(milr_result, testData$X, testData$ID, type = "instance") # predicted instance labels

# use auto-tuning (not run)
#milr_result <- milr(trainData$Z, trainData$X, trainData$ID, lambda = -1, numLambda = 20)
#coef(milr_result)      # coefficients
#fitted(milr_result)                    # fitted bag labels
#fitted(milr_result, type = "instance") # fitted instance labels
#summary(milr_result)   # summary milr
#predict(milr_result, testData$X, testData$ID)                    # predicted bag labels
#predict(milr_result, testData$X, testData$ID, type = "instance") # predicted instance labels

# use cv in auto-tuning (not run)
#milr_result <- milr(trainData$Z, trainData$X, trainData$ID, 
#                    lambda = -1, numLambda = 20, lambdaCriterion = "deviance")
#coef(milr_result)      # coefficients
#fitted(milr_result)                    # fitted bag labels
#fitted(milr_result, type = "instance") # fitted instance labels
#summary(milr_result)   # summary milr
#predict(milr_result, testData$X, testData$ID)                    # predicted bag labels
#predict(milr_result, testData$X, testData$ID, type = "instance") # predicted instance labels

Predict Method for milr Fits

Description

Predict Method for milr Fits

Usage

## S3 method for class 'milr'
predict(object, newdata = NULL, bag_newdata = NULL, type = "bag", ...)

Arguments

object

A fitted obejct of class inheriting from "milr".

newdata

Default is NULL. A matrix with variables to predict.

bag_newdata

Default is NULL. A vector. The labels of instances to bags. If newdata and bag_newdata both are NULL, return the fitted result.

type

The type of prediction required. Default is "bag", the predicted labels of bags. The "instance" option returns the predicted labels of instances.

...

further arguments passed to or from other methods.

Predict Method for softmax Fits

Description

Predict Method for softmax Fits

Usage

## S3 method for class 'softmax'
predict(object, newdata = NULL, bag_newdata = NULL, type = "bag", ...)

Arguments

object

A fitted obejct of class inheriting from "softmax".

newdata

Default is NULL. A matrix with variables to predict.

bag_newdata

Default is NULL. A vector. The labels of instances to bags. If newdata and bag_newdata both are NULL, return the fitted result.

type

The type of prediction required. Default is "bag", the predicted labels of bags. The "instance" option returns the predicted labels of instances.

...

further arguments passed to or from other methods.

Multiple-instance logistic regression via softmax function

Description

This function calculates the alternative maximum likelihood estimation for multiple-instance logistic regression through a softmax function (Xu and Frank, 2004; Ray and Craven, 2005).

Usage

softmax(y, x, bag, alpha = 0, ...)

Arguments

y

a vector. Bag-level binary labels.

x

the design matrix. The number of rows of x must be equal to the length of y.

bag

a vector, bag id.

alpha

A non-negative realnumber, the softmax parameter.

...

arguments to be passed to the optim function.

Value

a list including coefficients and fitted values.

References

S. Ray, and M. Craven. (2005) Supervised versus multiple instance learning: An empirical comparsion. in Proceedings of the 22nd International Conference on Machine Learnings, ACM, 697–704.
X. Xu, and E. Frank. (2004) Logistic regression and boosting for labeled bags of instances. in Advances in Knowledge Discovery and Data Mining, Springer, 272–281.

Examples

set.seed(100)
beta <- runif(10, -5, 5)
trainData <- DGP(40, 3, beta)
testData <- DGP(5, 3, beta)
# Fit softmax-MILR model S(0)
softmax_result <- softmax(trainData$Z, trainData$X, trainData$ID, alpha = 0)
coef(softmax_result)      # coefficients
fitted(softmax_result)                    # fitted bag labels
fitted(softmax_result, type = "instance") # fitted instance labels
predict(softmax_result, testData$X, testData$ID)                    # predicted bag labels
predict(softmax_result, testData$X, testData$ID, type = "instance") # predicted instance labels
# Fit softmax-MILR model S(3) (not run)
# softmax_result <- softmax(trainData$Z, trainData$X, trainData$ID, alpha = 3)