Skip to main content
Logistic Regression models the probability of each class using a linear combination of features. It’s fast, interpretable, and works well when the relationship between features and target is approximately linear.

​
Configuration

ParameterDescriptionDefault
CInverse of regularization strength. Smaller values = stronger regularization.1.0
SolverOptimization algorithm: lbfgs, liblinear, saga, sag, newton-cglbfgs
Max IterationsMaximum number of iterations for the solver to converge.1000
PenaltyRegularization type: l2, l1, elasticnet, nonel2
Random StateSeed for reproducibility.42

​
Input / Output

Type
InputX Train + Y Train
OutputTrained Model

​
When to use

Best for linearly separable data. Apply Scaling and Encoding before this component, as it is sensitive to feature magnitudes and requires numerical inputs.