Scribd
Upload
4 views1 page

Hyperparameters for Regression Models

The document outlines hyperparameters for Linear and Logistic Regression models, detailing their descriptions and default values. Key hyperparameters include fit_intercept, penalty type, and regularization strength. It provides essential information for configuring these models effectively.

Uploaded by

kingjaqfaq
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
4 views1 page

Hyperparameters for Regression Models

The document outlines hyperparameters for Linear and Logistic Regression models, detailing their descriptions and default values. Key hyperparameters include fit_intercept, penalty type, and regularization strength. It provides essential information for configuring these models effectively.

Uploaded by

kingjaqfaq
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Model Hyperparameters and Descriptions

Model Hyperparameter Description Default Value

Linear Regression fit_intercept Whether to calculate the intercept


True for this model.

Linear Regression normalize (deprecated) If True, the regressors X will


Falsebe normalized before regression.

Linear Regression copy_X If True, X will be copied; otherwise,


True it may be overwritten.

Linear Regression n_jobs The number of jobs to runNone


in parallel. -1 means using all processors.

Linear Regression positive If True, forces the coefficients


Falseto be positive.

Logistic Regression penalty Type of regularization (l1, 'l2'


l2, elasticnet, or none).

Logistic Regression dual Dual or primal formulationFalse


(only for l2 penalty and liblinear solver).

Logistic Regression tol Tolerance for stopping criteria.


1e-4

Logistic Regression C Inverse of regularization strength.


1.0 Smaller values specify stronger regularizat

Logistic Regression fit_intercept Whether to include an intercept


True in the model.

Logistic Regression intercept_scaling Scaling of the intercept term


1 when solver='liblinear'.

Logistic Regression class_weight Weights associated with classes


None (balanced or dictionary).

Logistic Regression random_state Seed for reproducibility. None

Logistic Regression solver Algorithm to use for optimization


'lbfgs' (lbfgs, liblinear, etc.).

Logistic Regression max_iter Maximum number of iterations


100 for solvers.

Logistic Regression multi_class Multi-class option (auto, ovr,


'auto'
or multinomial).

Logistic Regression n_jobs Number of CPU cores used


None
for parallelization (-1 to use all cores).

Logistic Regression l1_ratio Mixing parameter for ElasticNet


None regularization (only when penalty='elasticnet

You might also like