mlpack_linear_regression(1)

✨ NAME ✨

mlpack_linear_regression — bestie, it's literally simple linear regression and prediction 💖 (that's the whole thing, it's giving "tiny program, big brain")

---

🎀 SYNOPSIS 🎀

mlpack_linear_regression [-m unknown] [-l double] [-T unknown] [-t unknown] [-r unknown] [-V bool] [-M unknown] [-o unknown] [-h -v]

---

🌸 DESCRIPTION 🌸

okay so. this is an implementation of simple linear regression and simple ridge regression, using ordinary least squares. no cap.

it solves the problem

y = X * b + e

where X (which you specify with --training_file (-t)) and y (which you give either as the last column of the input matrix --training_file (-t), or via the --training_responses_file (-r) parameter) are known, and b is the variable you actually want. 👑

if the covariance matrix (X'X) is not invertible, or if your solution is overdetermined, then you gotta specify a Tikhonov regularization constant (with --lambda (-l)) that's greater than 0 — that regularizes the covariance matrix and makes it invertible again. problem solved, slay. 💅 the calculated b may be saved with the --output_predictions_file (-o) output parameter.

optionally, the calculated value of b gets used to predict the responses for another matrix X' (specified by the --test_file (-T) parameter):

y' = X' * b

and those predicted responses y' may be saved with the --output_predictions_file (-o) output parameter. also, fyi, this type of regression is related to least-angle regression, which mlpack implements as the lars program. 🦋

for example, to run a linear regression on the dataset X.csv with responses y.csv, saving the trained model to lr_model.bin, you could run:

$ mlpack_linear_regression --training_file X.csv --training_responses_file y.csv --output_model_file lr_model.bin

then, to use lr_model.bin to predict responses for a test set X_test.csv, saving the predictions to X_test_responses.csv, you'd run:

$ mlpack_linear_regression --input_model_file lr_model.bin --test_file X_test.csv --output_predictions_file X_test_responses.csv

like… it's that easy bestie. two commands and you're a data scientist. no thoughts, just regression. 😭

---

💜 OPTIONAL INPUT OPTIONS 💜

• --help (-h) [bool] — the default help info. the "what does this even do" button. 🫶
• --info [string] — prints help on one specific option. default value is ''.
• --input_model_file (-m) [unknown] — an existing LinearRegression model to use.
• --lambda (-l) [double] — Tikhonov regularization for ridge regression. if it's 0, the method reduces to linear regression. default value is 0. 🧚
• --test_file (-T) [unknown] — matrix containing X' (the test regressors).
• --training_file (-t) [unknown] — matrix containing the training set X (the regressors).
• --training_responses_file (-r) [unknown] — optional vector containing y (the responses). if you don't give it, then the responses are assumed to be the last row of the input file.
• --verbose (-v) [bool] — displays informational messages plus the full list of parameters and timers when it finishes running. main character energy, honestly. ✨
• --version (-V) [bool] — displays the version of mlpack.

---

🌈 OPTIONAL OUTPUT OPTIONS 🌈

• --output_model_file (-M) [unknown] — output LinearRegression model. your baby, saved to disk. 👑
• --output_predictions_file (-o) [unknown] — if --test_file is specified, this matrix is where the predicted responses get saved. 💖

---

⭐ ADDITIONAL INFORMATION ⭐

for further information — including relevant papers, citations, and theory — go check the documentation at http://www.mlpack.org, or the docs that came bundled with your distribution of mlpack. 🌸

---

and that's the whole page bestie, periodt. this was brought to you by mlpack-4.8.0, dated 22 June 2026 — go forth and regress, the vibes are immaculate, iykyk 🦋✨💅