Summary

The Lasso is a shrinkage and selection method for linear regression. Lasso4j is a Java implementation of the Lasso L1-constrained fitting for linear regression. This implementation is based on the glmnet R package which is implemented in Fortran.

Sample Output

• Input data: diabetes.data (source: http://www.stanford.edu/~hastie/Papers/LARS/diabetes.data)
• glmnet R package output: glmnet-diabetes-result.txt
• lasso4j output: lasso4j-output-for-diabetes.txt

Sample Usage

For using lasso4j, you need to download the jar file from Downloads section and add it to your classpath. For an example usage of the code, see this example.

Developed by Yasser Ganjisaffar