Package 's2net' reference manual

Title:	The Generalized Semi-Supervised Elastic-Net
Description:	Implements the generalized semi-supervised elastic-net. This method extends the supervised elastic-net problem, and thus it is a practical solution to the problem of feature selection in semi-supervised contexts. Its mathematical formulation is presented from a general perspective, covering a wide range of models. We focus on linear and logistic responses, but the implementation could be easily extended to other losses in generalized linear models. We develop a flexible and fast implementation, written in 'C++' using 'RcppArmadillo' and integrated into R via 'Rcpp' modules. See Culp, M. 2013 <doi:10.1080/10618600.2012.657139> for references on the Joint Trained Elastic-Net.
Authors:	Juan C. Laria [aut, cre] , Line H. Clemmensen [aut]
Maintainer:	Juan C. Laria <[email protected]>
License:	GPL (>= 2)
Version:	1.0.7
Built:	2024-11-27 06:31:49 UTC
Source:	CRAN

The Generalized Semi-Supervised Elastic-Net

Description

Implements the generalized semi-supervised elastic-net. This method extends the supervised elastic-net problem, and thus it is a practical solution to the problem of feature selection in semi-supervised contexts. Its mathematical formulation is presented from a general perspective, covering a wide range of models. We focus on linear and logistic responses, but the implementation could be easily extended to other losses in generalized linear models. We develop a flexible and fast implementation, written in 'C++' using 'RcppArmadillo' and integrated into R via 'Rcpp' modules. See Culp, M. 2013 <doi:10.1080/10618600.2012.657139> for references on the Joint Trained Elastic-Net.

Details

The DESCRIPTION file:

Package:	s2net
Type:	Package
Title:	The Generalized Semi-Supervised Elastic-Net
Version:	1.0.7
Date:	2024-03-04
Authors@R:	c(person("Juan C.", "Laria",, role = c("aut", "cre"), email = "[email protected]", comment = c(ORCID = "0000-0001-7734-9647")), person("Line H.", "Clemmensen",, role = c("aut"), email = "[email protected]"))
Description:	Implements the generalized semi-supervised elastic-net. This method extends the supervised elastic-net problem, and thus it is a practical solution to the problem of feature selection in semi-supervised contexts. Its mathematical formulation is presented from a general perspective, covering a wide range of models. We focus on linear and logistic responses, but the implementation could be easily extended to other losses in generalized linear models. We develop a flexible and fast implementation, written in 'C++' using 'RcppArmadillo' and integrated into R via 'Rcpp' modules. See Culp, M. 2013 <doi:10.1080/10618600.2012.657139> for references on the Joint Trained Elastic-Net.
License:	GPL (>= 2)
Imports:	Rcpp, methods, MASS
Depends:	stats
LinkingTo:	Rcpp, RcppArmadillo
Suggests:	knitr, rmarkdown, glmnet, Metrics, testthat
VignetteBuilder:	knitr
URL:	https://github.com/jlaria/s2net
BugReports:	https://github.com/jlaria/s2net/issues
Encoding:	UTF-8
RoxygenNote:	7.2.0
NeedsCompilation:	yes
Packaged:	2024-03-31 08:37:38 UTC; root
Author:	Juan C. Laria [aut, cre] (<https://orcid.org/0000-0001-7734-9647>), Line H. Clemmensen [aut]
Maintainer:	Juan C. Laria <[email protected]>
Repository:	CRAN
Date/Publication:	2024-03-31 10:30:02 UTC

Index of help topics:

Rcpp_s2net-class        Class 's2net'
auto_mpg                Auto MPG Data Set
predict.s2netR          S3 Methods for 's2netR' objects.
predict_Rcpp_s2net      Predict method for 's2net' C++ class.
print.s2Data            Print methods for S3 objects
s2Data                  Data wrapper for 's2net'.
s2Fista                 Hyper-parameter wrapper for FISTA.
s2Params                Hyper-parameter wrapper for 's2net'
s2net                   The Generalized Semi-Supervised Elastic-Net
s2netR                  Trains a generalized extended linear joint
                        trained model using semi-supervised data.
simulate_extra          Simulate extrapolated data
simulate_groups         Simulate data (two groups design)

Further information is available in the following vignettes:

`supervised`	The supervised `s2net` (source, pdf)

This package includes a very easy-to-use interface for handling data, with the s2Data function. The main function of the package is the s2netR function, which is a wrapper for the Rcpp_s2net (s2net) class.

Author(s)

Juan C. Laria [aut, cre] (<https://orcid.org/0000-0001-7734-9647>), Line H. Clemmensen [aut]

References

Laria, J.C., L. Clemmensen (2019). A generalized elastic-net for semi-supervised learning of sparse features.

Sogaard Larsen, J. et. al. (2019). Semi-supervised covariate shift modelling of spectroscopic data.

Ryan, K. J., & Culp, M. V. (2015). On semi-supervised linear regression in covariate shift problems. The Journal of Machine Learning Research, 16(1), 3183-3217.

Examples

data("auto_mpg")
train = s2Data(xL = auto_mpg$P1$xL, yL = auto_mpg$P1$yL,  xU = auto_mpg$P1$xU)

model = s2netR(train, 
                s2Params(lambda1 = 0.1, 
                           lambda2 = 0,
                           gamma1 = 0.1,
                           gamma2 = 100,
                           gamma3 = 0.1))

# here we tell it to transform the valid data as we did with train.
valid = s2Data(auto_mpg$P1$xU, auto_mpg$P1$yU, preprocess = train) 
ypred = predict(model, valid$xL)

## Not run: 
if(require(ggplot2)){
  ggplot() + 
    aes(x = ypred, y = valid$yL) + geom_point() + 
    geom_abline(intercept = 0, slope = 1, linetype = 2)
}

## End(Not run)
data("auto_mpg")
train = s2Data(xL = auto_mpg$P1$xL, yL = auto_mpg$P1$yL,  xU = auto_mpg$P1$xU)

model = s2netR(train, 
                s2Params(lambda1 = 0.1, 
                           lambda2 = 0,
                           gamma1 = 0.1,
                           gamma2 = 100,
                           gamma3 = 0.1))

# here we tell it to transform the valid data as we did with train.
valid = s2Data(auto_mpg$P1$xU, auto_mpg$P1$yU, preprocess = train) 
ypred = predict(model, valid$xL)

## Not run: 
if(require(ggplot2)){
  ggplot() + 
    aes(x = ypred, y = valid$yL) + geom_point() + 
    geom_abline(intercept = 0, slope = 1, linetype = 2)
}

## End(Not run)

Auto MPG Data Set

Description

This dataset was taken from the UCI Machine Learning Repository https://archive.ics.uci.edu/ml/datasets/Auto+MPG, and processed for the semi-supervised setting (Ryan and Culp, 2015).

Usage

data("auto_mpg")data("auto_mpg")

Format

There are two lists that contain partitions from a data frame with 398 observations on the following 9 variables.

mpg: a numeric vector
cylinders: an ordered factor with levels 3 < 4 < 5 < 6 < 8
displacement: a numeric vector
horsepower: a numeric vector
weight: a numeric vector
acceleration: a numeric vector
year: a numeric vector
origin: a factor

Details

This dataset is a slightly modified version of the dataset provided in the StatLib library. In line with the use by Ross Quinlan (1993) in predicting the attribute "mpg", 8 of the original instances were removed because they had unknown values for the "mpg" attribute. "The data concerns city-cycle fuel consumption in miles per gallon, to be predicted in terms of 3 multivalued discrete and 5 continuous attributes." (Quinlan, 1993)

Source

Quinlan,R. (1993). Combining Instance-Based and Model-Based Learning. In Proceedings on the Tenth International Conference of Machine Learning, 236-243, University of Massachusetts, Amherst. Morgan Kaufmann.

Dua, D. and Graff, C. (2019). UCI Machine Learning Repository [http://archive.ics.uci.edu/ml/]. Irvine, CA: University of California, School of Information and Computer Science.

References

Ryan, K. J., & Culp, M. V. (2015). On semi-supervised linear regression in covariate shift problems. The Journal of Machine Learning Research, 16(1), 3183-3217.

Examples

data(auto_mpg)
head(auto_mpg$P1$xL)
data(auto_mpg)
head(auto_mpg$P1$xL)

Predict method for `s2net` C++ class.

Description

This function provides an interface in R for the method predict in C++ class s2net.

Usage

predict_Rcpp_s2net(object, newX, type = "default")
predict_Rcpp_s2net(object, newX, type = "default")

Arguments

`object`	An object of class `Rcpp_s2net`.
`newX`	Data to make predictions. Could be a `s2Data` object (field `xL` is used) or a `matrix` (in the same space as the original data where the model was fitted).
`type`	Type of predictions. One of `"default"`: let the method figure it out; `"response"`: the linear predictor; `"probs"`: fitted probabilities; `class`: fitted class.

Details

This method is included as a high-level wrapper of object$predict().

Value

Returns a column matrix with the same number of rows/observations as newX.

Author(s)

Juan C. Laria

S3 Methods for `s2netR` objects.

Description

Generic predict method. Wrapper for the C++ class method s2net$predict.

Usage

## S3 method for class 's2netR'
predict(object, newX, type = "default", ...)
## S3 method for class 's2netR'
predict(object, newX, type = "default", ...)

Arguments

`object`	A `s2netR` object
`newX`	A matrix with the data to make predictions. It should be in the same scale as the original data. See `s2Data` to see how to format the data.
`type`	Type of predictions. One of `"default"` (figure it out from the train data), `"response", "probs", "class"`.
`...`	other parameters passed to predict

Value

A column matrix with predictions.

Examples

data("auto_mpg")
train = s2Data(xL = auto_mpg$P1$xL, yL = auto_mpg$P1$yL,  xU = auto_mpg$P1$xU)

model = s2netR(train, 
                s2Params(lambda1 = 0.1, 
                           lambda2 = 0,
                           gamma1 = 0.1,
                           gamma2 = 100,
                           gamma3 = 0.1),
                loss = "linear",
                frame = "ExtJT",
                proj = "auto",
                fista = s2Fista(5000, 1e-7, 1, 0.8))

valid = s2Data(auto_mpg$P1$xU, auto_mpg$P1$yU, preprocess = train)
ypred = predict(model, valid$xL)
## Not run: 
if(require(ggplot2)){
  ggplot() + 
    aes(x = ypred, y = valid$yL) + geom_point() + 
    geom_abline(intercept = 0, slope = 1, linetype = 2)
}

## End(Not run)
data("auto_mpg")
train = s2Data(xL = auto_mpg$P1$xL, yL = auto_mpg$P1$yL,  xU = auto_mpg$P1$xU)

model = s2netR(train, 
                s2Params(lambda1 = 0.1, 
                           lambda2 = 0,
                           gamma1 = 0.1,
                           gamma2 = 100,
                           gamma3 = 0.1),
                loss = "linear",
                frame = "ExtJT",
                proj = "auto",
                fista = s2Fista(5000, 1e-7, 1, 0.8))

valid = s2Data(auto_mpg$P1$xU, auto_mpg$P1$yU, preprocess = train)
ypred = predict(model, valid$xL)
## Not run: 
if(require(ggplot2)){
  ggplot() + 
    aes(x = ypred, y = valid$yL) + geom_point() + 
    geom_abline(intercept = 0, slope = 1, linetype = 2)
}

## End(Not run)

Print methods for S3 objects

Description

Very simple print methods to show basic information about these simple S3 objects.

Usage

## S3 method for class 's2Data'
print(x, ...)
## S3 method for class 's2Fista'
print(x, ...)
## S3 method for class 's2Data'
print(x, ...)
## S3 method for class 's2Fista'
print(x, ...)

Arguments

`x`	S3 object of class `s2Data` or `s2Fista`
`...`	other parameters passed to print

Class `s2net`

Description

This is the main class of this library, implemented in C++ and exposed to R using Rcpp modules. It can be used in R directly, although some generic S4 methods have been implemented to make it easier to interact in R.

Methods

predict: signature(object = "Rcpp_s2net"): See predict_Rcpp_s2net

Fields

beta:: Object of class matrix. The fitted model coefficients.
intercept:: The model intercept.

Class-Based Methods

initialize(data, loss):

data: s2Data object
loss: Loss function: 0 = linear, 1 = logit

setupFista(s2Fista):

Configures the FISTA internal algorithm.

predict(newX, type):

newX: New data matrix to make predictions.
type: 0 = default, 1 = response, 2 = probs, 3 = class

fit(params, frame, proj):

params: s2Params object
frame: 0 = "JT", 1 = "ExtJT"
proj: 0 = no, 1 = yes, 2 = auto

Author(s)

Juan C. Laria

Examples

data("auto_mpg")
train = s2Data(xL = auto_mpg$P1$xL, yL = auto_mpg$P1$yL,  xU = auto_mpg$P1$xU)

# We create the C++ object calling the new method (constructor)
obj = new(s2net, train, 0) # 0 = regression 
obj

# We call directly the $fit method of obj, 
obj$fit(s2Params(lambda1 = 0.01, 
                   lambda2 = 0.01, 
                   gamma1 = 0.05, 
                   gamma2 = 100, 
                   gamma3 = 0.05), 1, 2)
# fitted model
obj$beta

# We can test the results using the unlabeled data
test = s2Data(xL = auto_mpg$P1$xU, yL = auto_mpg$P1$yU,  preprocess = train)
ypred = obj$predict(test$xL, 0)

## Not run: 
if(require(ggplot2)){
  ggplot() + 
    aes(x = ypred, y = test$yL) + geom_point() + 
    geom_abline(intercept = 0, slope = 1, linetype = 2)
}

## End(Not run)

data("auto_mpg")
train = s2Data(xL = auto_mpg$P1$xL, yL = auto_mpg$P1$yL,  xU = auto_mpg$P1$xU)

# We create the C++ object calling the new method (constructor)
obj = new(s2net, train, 0) # 0 = regression 
obj

# We call directly the $fit method of obj, 
obj$fit(s2Params(lambda1 = 0.01, 
                   lambda2 = 0.01, 
                   gamma1 = 0.05, 
                   gamma2 = 100, 
                   gamma3 = 0.05), 1, 2)
# fitted model
obj$beta

# We can test the results using the unlabeled data
test = s2Data(xL = auto_mpg$P1$xU, yL = auto_mpg$P1$yU,  preprocess = train)
ypred = obj$predict(test$xL, 0)

## Not run: 
if(require(ggplot2)){
  ggplot() + 
    aes(x = ypred, y = test$yL) + geom_point() + 
    geom_abline(intercept = 0, slope = 1, linetype = 2)
}

## End(Not run)

Data wrapper for `s2net`.

Description

This function preprocess the data to fit a semi-supervised linear joint trained model.

Usage

s2Data(xL, yL, xU = NULL, preprocess = T)
s2Data(xL, yL, xU = NULL, preprocess = T)

Arguments

`xL`	The labeled data. Could be a `matrix` or `data.frame`.
`yL`	The labels associated with `xL`. Could be a `vector`, `matrix` or `data.frame`, of `factor` or `numeric` types.
`xU`	The unlabeled data (optional). Could be a `matrix` or `data.frame`.
`preprocess`	Should the input data be pre-processed? Possible values are: `TRUE` (default) The data is converted to a matrix. Factor variables are automatically coded using `model.matrix`. The data is scaled, and constant columns are removed. `FALSE` Do nothing. Keep in mind that the theoretical framework assumes that `xL` is centered. Unless you are absolutely sure, avoid this. Another object of class `s2Data` that was obtained from similar data (same original variables). This is useful when using train/validation sets, to apply the validation data the same transformation as train data.

Value

Returns an object of S3 class s2Data with fields

`xL`	Transformed labeled data
`yL`	Transformed labels. If `yL` was a factor, it is converted to numeric, and the base category is kept in `base`
`xU`	Tranformed unlabeled data
`type`	Type of task. This one is inferred from the response labels.
`base`	Base category for classification `0 = base`

In addition the following attributes are stored.

`pr:rm_cols`	logical vector of removed columns
`pr:center`	column center
`pr:scale`	column scale
`pr:ycenter`	yL center. Regression
`pr:yscale`	yL scale. Regression

Author(s)

Juan C. Laria

Examples

data("auto_mpg")

train = s2Data( xL = auto_mpg$P1$xL,
                  yL = auto_mpg$P1$yL,
                  xU = auto_mpg$P1$xU,
                  preprocess = TRUE )
show(train)

# Notice how ordered factor variable $cylinders is handled 
# .L (linear) .Q (quadratic) .C (cubic) and .^4
head(train$xL) 


#if you want to do validation with the unlabeled data
idx = sample(length(auto_mpg$P1$yU), 200)

train = s2Data(xL = auto_mpg$P1$xL, yL = auto_mpg$P1$yL, xU = auto_mpg$P1$xU[idx, ])

valid = s2Data(xL = auto_mpg$P1$xU[-idx, ], yL = auto_mpg$P1$yU[-idx], preprocess = train)

test = s2Data(xL = auto_mpg$P1$xU[idx, ], yL = auto_mpg$P1$yU[idx], preprocess = train)

train
valid
test
data("auto_mpg")

train = s2Data( xL = auto_mpg$P1$xL,
                  yL = auto_mpg$P1$yL,
                  xU = auto_mpg$P1$xU,
                  preprocess = TRUE )
show(train)

# Notice how ordered factor variable $cylinders is handled 
# .L (linear) .Q (quadratic) .C (cubic) and .^4
head(train$xL) 


#if you want to do validation with the unlabeled data
idx = sample(length(auto_mpg$P1$yU), 200)

train = s2Data(xL = auto_mpg$P1$xL, yL = auto_mpg$P1$yL, xU = auto_mpg$P1$xU[idx, ])

valid = s2Data(xL = auto_mpg$P1$xU[-idx, ], yL = auto_mpg$P1$yU[-idx], preprocess = train)

test = s2Data(xL = auto_mpg$P1$xU[idx, ], yL = auto_mpg$P1$yU[idx], preprocess = train)

train
valid
test

Hyper-parameter wrapper for FISTA.

Description

This is a very simple function that supplies the hyper-parameters for the Fast Iterative Soft-Threshold Algorithm (FISTA) that solves the s2net minimization problem.

Usage

s2Fista(MAX_ITER_INNER = 5000, TOL = 1e-07, t0 = 2, step = 0.1, use_warmstart = FALSE)
s2Fista(MAX_ITER_INNER = 5000, TOL = 1e-07, t0 = 2, step = 0.1, use_warmstart = FALSE)

Arguments

`MAX_ITER_INNER`	Number of iterations of FISTA
`TOL`	The relative tolerance. The algorith stops when the objective does not improve more than `TOL*`the null model's objective function evaluation, after two succesive iterations.
`t0`	The initial stepsize for backtracking.
`step`	The scale factor in the stepsize to backtrack until a valid step is found.
`use_warmstart`	Should we use a warm `beta` to fit the model? This is useful to speed-up hyper-parameter searching methods.

Value

Returns an object of S3 class s2Fista with the input arguments as fields.

References

Beck, A., & Teboulle, M. (2009). A fast iterative shrinkage-thresholding algorithm for linear inverse problems. SIAM journal on imaging sciences, 2(1), 183-202. doi:10.1137/080716542

Trains a generalized extended linear joint trained model using semi-supervised data.

Description

This function is a wrapper for the class s2net. It creates the C++ object and fits the model using input data.

Usage

s2netR(data, params, loss = "default", frame = "ExtJT", proj = "auto", 
        fista = NULL, S3 = TRUE)
s2netR(data, params, loss = "default", frame = "ExtJT", proj = "auto", 
        fista = NULL, S3 = TRUE)

Arguments

`data`	A `s2Data` object with the (training) data.
`params`	A `s2Params` object with the model hyper-parameters.
`loss`	Loss function. One of `"default"` (figure it out from the data), `"linear"` or `"logit"`.
`frame`	The semi-supervised frame: `"ExtJT"` (the extended linear joint trained model), `"JT"` (the linear joint trained model from Ryan and Culp. 2015)
`proj`	Should the unlabeled data be shifted to remove the model's effect? One of `"no", "yes", "auto"` (option `auto` shifts the unlabeled data if the angle betwen beta and the center of the data is important)
`fista`	Fista setup parameters. An object of class `s2Fista`.
`S3`	Boolean: should the method return an S3 object (default) or a C++ object?

Value

Returns an object of S3 class s2netR or a C++ object of class s2net

Author(s)

Juan C. Laria

References

Ryan, K. J., & Culp, M. V. (2015). On semi-supervised linear regression in covariate shift problems. The Journal of Machine Learning Research, 16(1), 3183-3217.

Examples

data("auto_mpg")
train = s2Data(xL = auto_mpg$P1$xL, yL = auto_mpg$P1$yL,  xU = auto_mpg$P1$xU)

model = s2netR(train, 
                s2Params(lambda1 = 0.1, 
                           lambda2 = 0,
                           gamma1 = 0.1,
                           gamma2 = 100,
                           gamma3 = 0.1),
                loss = "linear",
                frame = "ExtJT",
                proj = "auto",
                fista = s2Fista(5000, 1e-7, 1, 0.8))

valid = s2Data(auto_mpg$P1$xU, auto_mpg$P1$yU, preprocess = train)
ypred = predict(model, valid$xL)

## Not run: 
if(require(ggplot2)){
  ggplot() + 
    aes(x = ypred, y = valid$yL) + geom_point() + 
    geom_abline(intercept = 0, slope = 1, linetype = 2)
}

## End(Not run)

data("auto_mpg")
train = s2Data(xL = auto_mpg$P1$xL, yL = auto_mpg$P1$yL,  xU = auto_mpg$P1$xU)

model = s2netR(train, 
                s2Params(lambda1 = 0.1, 
                           lambda2 = 0,
                           gamma1 = 0.1,
                           gamma2 = 100,
                           gamma3 = 0.1),
                loss = "linear",
                frame = "ExtJT",
                proj = "auto",
                fista = s2Fista(5000, 1e-7, 1, 0.8))

valid = s2Data(auto_mpg$P1$xU, auto_mpg$P1$yU, preprocess = train)
ypred = predict(model, valid$xL)

## Not run: 
if(require(ggplot2)){
  ggplot() + 
    aes(x = ypred, y = valid$yL) + geom_point() + 
    geom_abline(intercept = 0, slope = 1, linetype = 2)
}

## End(Not run)

Hyper-parameter wrapper for `s2net`

Description

This is a very simple function that collapses the input parameters into a named vector to supply to C++ methods.

Usage

s2Params(lambda1, lambda2 = 0, gamma1 = 0, gamma2 = 0, gamma3 = 0)
s2Params(lambda1, lambda2 = 0, gamma1 = 0, gamma2 = 0, gamma3 = 0)

Arguments

`lambda1`	elastic-net regularization parameter - $l_1$ norm.
`lambda2`	elastic-net regularization parameter - $l_2$ norm.
`gamma1`	s2net weight hyper-parameter.
`gamma2`	s2net covariance hyper-parameter (between 1 and `Inf`).
`gamma3`	s2net shift hyper-parameter (between 0 and 1).

Value

Returns a named vector of S3 class s2Params.

Simulate extrapolated data

Description

Simulated data scenarios described in the paper from Ryan and Culp (2015).

Usage

simulate_extra(n_source = 100, n_target = 100, p = 1000, shift = 10, 
               scenario = "same", response = "linear", sigma2 = 2.5)
simulate_extra(n_source = 100, n_target = 100, p = 1000, shift = 10, 
               scenario = "same", response = "linear", sigma2 = 2.5)

Arguments

`n_source`	Number of source samples (labeled)
`n_target`	Number of target samples (unlabeled)
`p`	Number of variables ( `p > 10`)
`shift`	The shift applied to the first 10 columns of xU.
`scenario`	Simulation scenario. One of `"same"` (same distribution), `"lucky"` (extrapolation with lucky $\beta$ ), `"unlucky"` (extrapolation with unlucky $\beta$ )
`response`	Type of response: `"linear"` or `"logit"`
`sigma2`	The variance of the error term, linear response case.

Value

A list, with

xL: data frame with the labeled (source) data
yL: labels associated with xL
xU: data frame with the unlabeled (target) data
yU: labels associated with xU (for validation/testing)

References

Ryan, K. J., & Culp, M. V. (2015). On semi-supervised linear regression in covariate shift problems. The Journal of Machine Learning Research, 16(1), 3183-3217.

Examples

set.seed(0)
data = simulate_extra()

train = s2Data(data$xL, data$yL, data$xU)
valid = s2Data(data$xU, data$yU, preprocess = train)

model = s2netR(train, s2Params(0.1))
ypred = predict(model, valid$xL)
plot(ypred, valid$yL)
set.seed(0)
data = simulate_extra()

train = s2Data(data$xL, data$yL, data$xU)
valid = s2Data(data$xU, data$yU, preprocess = train)

model = s2netR(train, s2Params(0.1))
ypred = predict(model, valid$xL)
plot(ypred, valid$yL)

Simulate data (two groups design)

Description

Simulated data scenario described in paper [citation here].

Usage

simulate_groups(n_source = 100, n_target = 100, p = 200, response = "linear")
simulate_groups(n_source = 100, n_target = 100, p = 200, response = "linear")

Arguments

`n_source`	Number of labeled observations
`n_target`	Number of unlabeled (target) observations
`p`	Number of variables
`response`	Type of response: `"linear"` or `"logit"`

Value

A list, with

xL: data frame with the labeled (source) data
yL: labels associated with xL
xU: data frame with the unlabeled (target) data
yU: labels associated with xU (for validation/testing)

Author(s)

Juan C. Laria

Package 's2net'

Help Index

The Generalized Semi-Supervised Elastic-Net

Description

Details

Author(s)

References

See Also

Examples

Auto MPG Data Set

Description

Usage

Format

Details

Source

References

Examples

Predict method for s2net C++ class.

Description

Usage

Arguments

Details

Value

Author(s)

See Also

S3 Methods for s2netR objects.

Description

Usage

Arguments

Value

See Also

Examples

Print methods for S3 objects

Description

Usage

Arguments

See Also

Class s2net

Description

Methods

Fields

Class-Based Methods

Author(s)

Examples

Data wrapper for s2net.

Description

Usage

Arguments

Value

Author(s)

See Also

Examples

Hyper-parameter wrapper for FISTA.

Description

Usage

Arguments

Value

References

See Also

Trains a generalized extended linear joint trained model using semi-supervised data.

Description

Usage

Arguments

Value

Author(s)

References

See Also

Examples

Hyper-parameter wrapper for s2net

Description

Usage

Arguments

Value

See Also

Simulate extrapolated data

Description

Usage

Arguments

Value

References

Predict method for `s2net` C++ class.

S3 Methods for `s2netR` objects.

Class `s2net`

Data wrapper for `s2net`.

Hyper-parameter wrapper for `s2net`