This is documentation for Orange 2.7. For the latest documentation, see Orange 3.

Linear Regression Learner

Linear Regression

Learns a linear function of its input data.


  • Data (Table)

    Input data table

  • Learner

    The learning algorithm with the supplied parameters

  • Predictor

    Trained regressor

  • Model Statisics

    A data table containing trained model statistics

Signal Predictor and Model Statistics send the output signal only if input signal Data is present.


Linear Regression widget construct a learner/predictor that learns a linear function from its input data. Furthermore Lasso and Ridge regularization parameters can be specified.

Linear Regression interface
  1. The learner/predictor name
  2. Train an ordinary least squares or ridge regression model
  3. If Ridge lambda is checked the learner will build a ridge regression model with 4 as the lambda parameter.
  4. Ridge lambda parameter.
  5. Use Lasso regularization.
  6. The Lasso bound (bound on the beta vector L1 norm)
  7. Tolerance (any beta value lower then this will be forced to 0)