mlpack_lars(26 December 2016) | mlpack_lars(26 December 2016) |
NAME¶
mlpack_lars - larsSYNOPSIS¶
mlpack_lars [-h] [-v]
DESCRIPTION¶
An implementation of LARS: Least Angle Regression (Stagewise/laSso). This is a stage-wise homotopy-based algorithm for L1-regularized linear regression (LASSO) and L1+L2-regularized linear regression (Elastic Net).This program is able to train a LARS/LASSO/Elastic Net model or load a model from file, output regression predictions for a test set, and save the trained model to a file. The LARS algorithm is described in more detail below:
Let X be a matrix where each row is a point and each column is a dimension, and let y be a vector of targets.
The Elastic Net problem is to solve
min_beta 0.5 || X * beta - y ||_2^2 + lambda_1 ||beta||_1 + 0.5 lambda_2 ||beta||_2^2If --lambda1 > 0 and --lambda2 = 0, the problem is the LASSO. If --lambda1 > 0 and --lambda2 > 0, the problem is the Elastic Net. If --lambda1 = 0 and --lambda2 > 0, the problem is ridge regression. If --lambda1 = 0 and --lambda2 = 0, the problem is unregularized linear regression.
For efficiency reasons, it is not recommended to use this algorithm with --lambda_1 = 0. In that case, use the 'linear_regression' program, which implements both unregularized linear regression and ridge regression.
To train a LARS/LASSO/Elastic Net model, the --input_file and --responses_file parameters must be given. The --lambda1 --lambda2, and --use_cholesky arguments control the training parameters. A trained model can be saved with the --output_model_file, or, if training is not desired at all, a model can be loaded with --input_model_file. Any output predictions from a test file can be saved into the file specified by the --output_predictions option.
OPTIONAL INPUT OPTIONS¶
- --help (-h)
- Default help info.
- --info [string]
- Get help on a specific module or option. Default value ''.
- --input_file (-i) [string]
- File containing covariates (X). Default value ’'. --input_model_file (-m) [string] File to load model from. Default value ''.
- --lambda1 (-l) [double]
- Regularization parameter for l1-norm penalty. Default value 0.
- --lambda2 (-L) [double]
- Regularization parameter for l2-norm penalty. Default value 0. --responses_file (-r) [string] File containing y (responses/observations). Default value ''.
- --test_file (-t) [string]
- File containing points to regress on (test points). Default value ''.
- --use_cholesky (-c)
- Use Cholesky decomposition during computation rather than explicitly computing the full Gram matrix.
- --verbose (-v)
- Display informational messages and the full list of parameters and timers at the end of execution.
- --version (-V)
- Display the version of mlpack.