@@ -496,15 +496,15 @@ class PoissonRegressor(_GeneralizedLinearRegressor):
496496 Parameters
497497 ----------
498498 alpha : float, default=1
499- Constant that multiplies the penalty term and thus determines the
499+ Constant that multiplies the L2 penalty term and determines the
500500 regularization strength. ``alpha = 0`` is equivalent to unpenalized
501501 GLMs. In this case, the design matrix `X` must have full column rank
502502 (no collinearities).
503- Values must be in the range `[0.0, inf)`.
503+ Values of `alpha` must be in the range `[0.0, inf)`.
504504
505505 fit_intercept : bool, default=True
506506 Specifies if a constant (a.k.a. bias or intercept) should be
507- added to the linear predictor (X @ coef + intercept).
507+ added to the linear predictor (` X @ coef + intercept` ).
508508
509509 solver : {'lbfgs', 'newton-cholesky'}, default='lbfgs'
510510 Algorithm to use in the optimization problem:
@@ -780,15 +780,15 @@ class TweedieRegressor(_GeneralizedLinearRegressor):
780780 For ``0 < power < 1``, no distribution exists.
781781
782782 alpha : float, default=1
783- Constant that multiplies the penalty term and thus determines the
783+ Constant that multiplies the L2 penalty term and determines the
784784 regularization strength. ``alpha = 0`` is equivalent to unpenalized
785785 GLMs. In this case, the design matrix `X` must have full column rank
786786 (no collinearities).
787- Values must be in the range `[0.0, inf)`.
787+ Values of `alpha` must be in the range `[0.0, inf)`.
788788
789789 fit_intercept : bool, default=True
790790 Specifies if a constant (a.k.a. bias or intercept) should be
791- added to the linear predictor (X @ coef + intercept).
791+ added to the linear predictor (` X @ coef + intercept` ).
792792
793793 link : {'auto', 'identity', 'log'}, default='auto'
794794 The link function of the GLM, i.e. mapping from linear predictor
0 commit comments