issue#290 integrate upstream changes

Author: floriankozikowski

.github/workflows/flake8.yml

@@ -30,4 +30,4 @@ jobs:
       - name: Check doc style with pydocstyle
         run: |
           pip install pydocstyle
-          pydocstyle skglm --ignore='D100',D102,'D104','D107','D203','D213','D413'
+          pydocstyle skglm --ignore='D100',D102,'D104','D105','D107','D203','D213','D413',

skglm/estimators.py

@@ -449,6 +449,11 @@ def path(self, X, y, alphas, coef_init=None, return_n_iter=True, **params):
             warm_start=self.warm_start, verbose=self.verbose)
         return solver.path(X, y, datafit, penalty, alphas, coef_init, return_n_iter)
 
+    def __sklearn_tags__(self):
+        tags = super().__sklearn_tags__()
+        tags.input_tags.sparse = True
+        return tags
+
 
 class WeightedLasso(RegressorMixin, LinearModel):
     r"""WeightedLasso estimator based on Celer solver and primal extrapolation.
@@ -612,6 +617,11 @@ def fit(self, X, y):
             warm_start=self.warm_start, verbose=self.verbose)
         return _glm_fit(X, y, self, Quadratic(), penalty, solver)
 
+    def __sklearn_tags__(self):
+        tags = super().__sklearn_tags__()
+        tags.input_tags.sparse = True
+        return tags
+
 
 class ElasticNet(RegressorMixin, LinearModel):
     r"""Elastic net estimator.
@@ -766,6 +776,11 @@ def fit(self, X, y):
         return _glm_fit(X, y, self, Quadratic(),
                         L1_plus_L2(self.alpha, self.l1_ratio, self.positive), solver)
 
+    def __sklearn_tags__(self):
+        tags = super().__sklearn_tags__()
+        tags.input_tags.sparse = True
+        return tags
+
 
 class MCPRegression(RegressorMixin, LinearModel):
     r"""Linear regression with MCP penalty estimator.
@@ -954,6 +969,11 @@ def fit(self, X, y):
             warm_start=self.warm_start, verbose=self.verbose)
         return _glm_fit(X, y, self, Quadratic(), penalty, solver)
 
+    def __sklearn_tags__(self):
+        tags = super().__sklearn_tags__()
+        tags.input_tags.sparse = True
+        return tags
+
 
 class SparseLogisticRegression(LinearClassifierMixin, SparseCoefMixin, BaseEstimator):
     r"""Sparse Logistic regression estimator.
@@ -1381,6 +1401,11 @@ def fit(self, X, y):
 
         return self
 
+    def __sklearn_tags__(self):
+        tags = super().__sklearn_tags__()
+        tags.input_tags.sparse = True
+        return tags
+
 
 class MultiTaskLasso(RegressorMixin, LinearModel):
     r"""MultiTaskLasso estimator.