msgl: High Dimensional Multiclass Classification Using Sparse Group Lasso

Share:

Multinomial logistic regression with sparse group lasso penalty. Simultaneous feature selection and parameter estimation for classification. Suitable for high dimensional multiclass classification with many classes. The algorithm computes the sparse group lasso penalized maximum likelihood estimate. Use of parallel computing for cross validation and subsampling is supported through the 'foreach' and 'doParallel' packages. Development version is on GitHub, please report package issues on GitHub.

Author
Martin Vincent
Date of publication
2016-09-28 19:35:07
Maintainer
Martin Vincent <martin.vincent.dk@gmail.com>
License
GPL (>= 2)
Version
2.3.0
URLs

View on CRAN

Man pages

best_model.msgl
Index of best model
coef.msgl
Extract nonzero coefficients
Err.msgl
Compute error rates
features.msgl
Nonzero features
features_stat.msgl
Extract feature statistics
models.msgl
Extract the fitted models
msgl
Fit a multinomial sparse group lasso regularization path.
msgl.algorithm.config
Create a new algorithm configuration
msgl.cv
Multinomial sparse group lasso cross validation
msgl.lambda.seq
Computes a lambda sequence for the regularization path
msgl.standard.config
Standard msgl algorithm configuration
msgl.subsampling
Multinomial sparse group lasso generic subsampling procedure
nmod.msgl
Returns the number of models in a msgl object
parameters.msgl
Nonzero parameters
parameters_stat.msgl
Extracting parameter statistics
predict.msgl
Predict
print.msgl
Print function for msgl
sim.data
Simulated data set

Files in this package

msgl
msgl/inst
msgl/inst/CITATION
msgl/inst/NEWS.Rd
msgl/inst/doc
msgl/inst/doc/quick-start.html
msgl/inst/doc/quick-start.Rmd
msgl/tests
msgl/tests/msgl_grouping_test_3.R
msgl/tests/msgl_test_1.R
msgl/tests/msgl_grouping_test_1.R
msgl/tests/msgl_test_3.R
msgl/tests/msgl_test_4.R
msgl/tests/msgl_cv_test_1.R
msgl/tests/msgl_grouping_test_2.R
msgl/tests/msgl_grouping_test_4.R
msgl/tests/msgl_cv_test_3.R
msgl/tests/msgl_configuration_test.R
msgl/tests/msgl_predict_test_1.R
msgl/tests/msgl_predict_test_2.R
msgl/tests/msgl_cv_test_4.R
msgl/tests/msgl_sub_test_1.R
msgl/tests/msgl_test_2.R
msgl/tests/msgl_cv_test_2.R
msgl/src
msgl/src/Makevars
msgl/src/multinomial_response.h
msgl/src/multinomial_loss.h
msgl/src/rmsgl.cpp
msgl/NAMESPACE
msgl/data
msgl/data/SimData.RData
msgl/R
msgl/R/msgl_fit.R
msgl/R/msgl_cv.R
msgl/R/msgl_config.R
msgl/R/msgl_subsampling.R
msgl/R/msgl_navigate.R
msgl/R/startup.R
msgl/R/msgl_predict.R
msgl/vignettes
msgl/vignettes/quick-start.Rmd
msgl/MD5
msgl/build
msgl/build/vignette.rds
msgl/DESCRIPTION
msgl/man
msgl/man/msgl.standard.config.Rd
msgl/man/features.msgl.Rd
msgl/man/msgl.subsampling.Rd
msgl/man/print.msgl.Rd
msgl/man/coef.msgl.Rd
msgl/man/best_model.msgl.Rd
msgl/man/predict.msgl.Rd
msgl/man/models.msgl.Rd
msgl/man/Err.msgl.Rd
msgl/man/msgl.Rd
msgl/man/parameters.msgl.Rd
msgl/man/msgl.cv.Rd
msgl/man/sim.data.Rd
msgl/man/features_stat.msgl.Rd
msgl/man/msgl.lambda.seq.Rd
msgl/man/parameters_stat.msgl.Rd
msgl/man/msgl.algorithm.config.Rd
msgl/man/nmod.msgl.Rd