Merge pull request #504 from sebp/sklearn-1-6

Add support for scikit-learn 1.6

.pre-commit-config.yaml

@@ -43,6 +43,6 @@ repos:
     hooks:
     -   id: black-jupyter
 -   repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.7.1
+    rev: v0.8.4
     hooks:
     -   id: ruff

README.rst

@@ -39,7 +39,7 @@ Requirements
 - numpy
 - osqp
 - pandas 1.4.0 or later
-- scikit-learn 1.4 or 1.5
+- scikit-learn 1.6
 - scipy
 - C/C++ compiler
 

ci/appveyor/py310.ps1

@@ -1,4 +1,4 @@
 $env:CI_PYTHON_VERSION="3.10.*"
 $env:CI_PANDAS_VERSION="1.5.*"
 $env:CI_NUMPY_VERSION="1.25.*"
-$env:CI_SKLEARN_VERSION="1.4.*"
+$env:CI_SKLEARN_VERSION="1.6.*"

ci/appveyor/py311.ps1

@@ -1,4 +1,4 @@
 $env:CI_PYTHON_VERSION="3.11.*"
 $env:CI_PANDAS_VERSION="2.0.*"
 $env:CI_NUMPY_VERSION="1.26.*"
-$env:CI_SKLEARN_VERSION="1.5.*"
+$env:CI_SKLEARN_VERSION="1.6.*"

ci/appveyor/py312.ps1

@@ -1,4 +1,4 @@
 $env:CI_PYTHON_VERSION="3.12.*"
 $env:CI_PANDAS_VERSION="2.2.*"
 $env:CI_NUMPY_VERSION="2.0.*"
-$env:CI_SKLEARN_VERSION="1.5.*"
+$env:CI_SKLEARN_VERSION="1.6.*"

ci/appveyor/py313.ps1

@@ -1,4 +1,4 @@
 $env:CI_PYTHON_VERSION="3.13.*"
 $env:CI_PANDAS_VERSION="2.2.*"
 $env:CI_NUMPY_VERSION="2.1.*"
-$env:CI_SKLEARN_VERSION="1.5.*"
+$env:CI_SKLEARN_VERSION="1.6.*"

ci/deps/py310.sh

@@ -2,5 +2,5 @@
 export CI_PYTHON_VERSION='3.10.*'
 export CI_PANDAS_VERSION='1.5.*'
 export CI_NUMPY_VERSION='1.25.*'
-export CI_SKLEARN_VERSION='1.4.*'
+export CI_SKLEARN_VERSION='1.6.*'
 export CI_NO_SLOW=false

ci/deps/py311.sh

@@ -2,5 +2,5 @@
 export CI_PYTHON_VERSION='3.11.*'
 export CI_PANDAS_VERSION='2.0.*'
 export CI_NUMPY_VERSION='1.26.*'
-export CI_SKLEARN_VERSION='1.5.*'
+export CI_SKLEARN_VERSION='1.6.*'
 export CI_NO_SLOW=true

ci/deps/py312.sh

@@ -2,5 +2,5 @@
 export CI_PYTHON_VERSION='3.12.*'
 export CI_PANDAS_VERSION='2.2.*'
 export CI_NUMPY_VERSION='2.0.*'
-export CI_SKLEARN_VERSION='1.5.*'
+export CI_SKLEARN_VERSION='1.6.*'
 export CI_NO_SLOW=true

ci/deps/py313.sh

@@ -2,5 +2,5 @@
 export CI_PYTHON_VERSION='3.13.*'
 export CI_PANDAS_VERSION='2.2.*'
 export CI_NUMPY_VERSION='2.1.*'
-export CI_SKLEARN_VERSION='1.5.*'
+export CI_SKLEARN_VERSION='1.6.*'
 export CI_NO_SLOW=false

doc/conf.py

@@ -217,7 +217,7 @@
 }
 
 intersphinx_mapping = {
-    "sklearn": ("https://scikit-learn.org/1.5", None),
+    "sklearn": ("https://scikit-learn.org/1.6", None),
     "cython": ("https://cython.readthedocs.io/en/latest/", None),
     "scipy": ("https://docs.scipy.org/doc/scipy/", None),
     "pandas": ("https://pandas.pydata.org/docs/", None),

doc/install.rst

@@ -100,6 +100,6 @@ The current minimum dependencies to run scikit-survival are:
 - numpy
 - osqp
 - pandas 1.4.0 or later
-- scikit-learn 1.4 or 1.5
+- scikit-learn 1.6
 - scipy
 - C/C++ compiler

pyproject.toml

@@ -9,8 +9,7 @@ requires = [
     "numpy>=2.0.0",
 
     # scikit-learn requirements
-    "scikit-learn~=1.4.0; python_version<='3.12'",
-    "scikit-learn~=1.5.0; python_version=='3.13'",
+    "scikit-learn~=1.6.1; python_version<='3.13'",
     "scikit-learn; python_version>'3.13'",
 ]
 build-backend = "setuptools.build_meta"
@@ -51,7 +50,7 @@ dependencies = [
     "osqp !=0.6.0,!=0.6.1",
     "pandas >=1.4.0",
     "scipy >=1.3.2",
-    "scikit-learn >=1.4.0,<1.6",
+    "scikit-learn >=1.6.1,<1.7",
 ]
 dynamic = ["version"]
 
@@ -188,13 +187,21 @@ target-version = "py310"
 ignore = ["C408"]
 ignore-init-module-imports = true
 select = [
-    "C4",
-    "C9",
+    # pycodestyle
     "E",
+    "W",
+    # mccabe
+    "C90",
+    # pyflakes
     "F",
+    # isort
     "I",
+    # flake8-builtins
+    "A",
+    # flake8-comprehensions
+    "C4",
+    # flake8-pytest-style
     "PT",
-    "W",
 ]
 
 [tool.ruff.lint.flake8-pytest-style]

sksurv/__init__.py

@@ -44,7 +44,7 @@ def show_versions():
         max(map(len, deps)),
         max(map(len, sys_info.keys())),
     )
-    fmt = "{0:<%ds}: {1}" % minwidth
+    fmt = f"{{0:<{minwidth}s}}: {{1}}"
 
     print("SYSTEM")
     print("------")

sksurv/base.py

@@ -99,5 +99,7 @@ def score(self, X, y):
         result = concordance_index_censored(y[name_event], y[name_time], risk_score)
         return result[0]
 
-    def _more_tags(self):
-        return {"requires_y": True}
+    def __sklearn_tags__(self):
+        tags = super().__sklearn_tags__()
+        tags.target_tags.required = True
+        return tags

sksurv/bintrees/_binarytrees.pyx

@@ -10,6 +10,7 @@
 #
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
+cimport cython
 from libcpp cimport bool
 from libcpp.cast cimport dynamic_cast
 
@@ -76,18 +77,21 @@ cdef class BaseTree:
         return self.count_larger(key)
 
 
+@cython.final
 cdef class RBTree(BaseTree):
     def __cinit__(self, int size):
         if size <= 0:
             raise ValueError('size must be greater zero')
         self.treeptr = new rbtree(size)
 
+@cython.final
 cdef class AVLTree(BaseTree):
     def __cinit__(self, int size):
         if size <= 0:
             raise ValueError('size must be greater zero')
         self.treeptr = dynamic_cast[rbtree_ptr](new avl(size))
 
+@cython.final
 cdef class AATree(BaseTree):
     def __cinit__(self, int size):
         if size <= 0:

sksurv/ensemble/boosting.py

@@ -21,10 +21,15 @@
 from sklearn.model_selection import train_test_split
 from sklearn.tree import DecisionTreeRegressor
 from sklearn.tree._tree import DTYPE
-from sklearn.utils import check_random_state
 from sklearn.utils._param_validation import Interval, StrOptions
 from sklearn.utils.extmath import squared_norm
-from sklearn.utils.validation import _check_sample_weight, check_array, check_is_fitted
+from sklearn.utils.validation import (
+    _check_sample_weight,
+    check_array,
+    check_is_fitted,
+    check_random_state,
+    validate_data,
+)
 
 from ..base import SurvivalAnalysisMixin
 from ..linear_model.coxph import BreslowEstimator
@@ -389,7 +394,7 @@ def fit(self, X, y, sample_weight=None):
         if not self.warm_start:
             self._clear_state()
 
-        X = self._validate_data(X, ensure_min_samples=2)
+        X = validate_data(self, X, ensure_min_samples=2)
         event, time = check_array_survival(X, y)
 
         sample_weight = _check_sample_weight(sample_weight, X)
@@ -398,7 +403,7 @@ def fit(self, X, y, sample_weight=None):
         Xi = np.column_stack((np.ones(n_samples), X))
 
         self._loss = LOSS_FUNCTIONS[self.loss]()
-        if isinstance(self._loss, (CensoredSquaredLoss, IPCWLeastSquaresError)):
+        if isinstance(self._loss, CensoredSquaredLoss | IPCWLeastSquaresError):
             time = np.log(time)
 
         if not self._is_fitted():
@@ -470,7 +475,7 @@ def predict(self, X):
             Predicted risk scores.
         """
         check_is_fitted(self, "estimators_")
-        X = self._validate_data(X, reset=False)
+        X = validate_data(self, X, reset=False)
 
         return self._predict(X)
 
@@ -957,7 +962,7 @@ def _set_max_features(self):
                 max_features = max(1, int(np.log2(self.n_features_in_)))
         elif self.max_features is None:
             max_features = self.n_features_in_
-        elif isinstance(self.max_features, (numbers.Integral, np.integer)):
+        elif isinstance(self.max_features, numbers.Integral):
             max_features = self.max_features
         else:  # float
             max_features = max(1, int(self.max_features * self.n_features_in_))
@@ -1234,7 +1239,8 @@ def fit(self, X, y, sample_weight=None, monitor=None):
         if not self.warm_start:
             self._clear_state()
 
-        X = self._validate_data(
+        X = validate_data(
+            self,
             X,
             ensure_min_samples=2,
             order="C",
@@ -1256,7 +1262,7 @@ def fit(self, X, y, sample_weight=None, monitor=None):
         # self.loss is guaranteed to be a string
         self._loss = self._get_loss(sample_weight=sample_weight)
 
-        if isinstance(self._loss, (CensoredSquaredLoss, IPCWLeastSquaresError)):
+        if isinstance(self._loss, CensoredSquaredLoss | IPCWLeastSquaresError):
             time = np.log(time)
 
         if self.n_iter_no_change is not None:
@@ -1315,13 +1321,13 @@ def fit(self, X, y, sample_weight=None, monitor=None):
             begin_at_stage = self.estimators_.shape[0]
             # The requirements of _raw_predict
             # are more constrained than fit. It accepts only CSR
-            # matrices. Finite values have already been checked in _validate_data.
+            # matrices. Finite values have already been checked in validate_data.
             X_train = check_array(
                 X_train,
                 dtype=DTYPE,
                 order="C",
                 accept_sparse="csr",
-                force_all_finite=False,
+                ensure_all_finite=False,
             )
             raw_predictions = self._raw_predict(X_train)
             self._resize_state()
@@ -1390,7 +1396,7 @@ def _dropout_raw_predict(self, X):
         return raw_predictions
 
     def _dropout_staged_raw_predict(self, X):
-        X = self._validate_data(X, dtype=DTYPE, order="C", accept_sparse="csr")
+        X = validate_data(self, X, dtype=DTYPE, order="C", accept_sparse="csr")
         raw_predictions = self._raw_predict_init(X)
 
         n_estimators, K = self.estimators_.shape
@@ -1438,7 +1444,7 @@ def predict(self, X):
         """
         check_is_fitted(self, "estimators_")
 
-        X = self._validate_data(X, reset=False, order="C", accept_sparse="csr", dtype=DTYPE)
+        X = validate_data(self, X, reset=False, order="C", accept_sparse="csr", dtype=DTYPE)
         return self._predict(X)
 
     def staged_predict(self, X):

sksurv/ensemble/forest.py

@@ -14,8 +14,8 @@
     _parallel_build_trees,
 )
 from sklearn.tree._tree import DTYPE
-from sklearn.utils._tags import _safe_tags
-from sklearn.utils.validation import check_is_fitted, check_random_state
+from sklearn.utils._tags import get_tags
+from sklearn.utils.validation import check_is_fitted, check_random_state, validate_data
 
 from ..base import SurvivalAnalysisMixin
 from ..metrics import concordance_index_censored
@@ -29,18 +29,20 @@
 MAX_INT = np.iinfo(np.int32).max
 
 
-def _more_tags_patch(self):
-    # BaseForest._more_tags calls
+def _sklearn_tags_patch(self):
+    # BaseForest.__sklearn_tags__ calls
     # type(self.estimator)(criterion=self.criterions),
     # which is incompatible with LogrankCriterion
     if isinstance(self, _BaseSurvivalForest):
         estimator = type(self.estimator)()
     else:
         estimator = type(self.estimator)(criterion=self.criterion)
-    return {"allow_nan": _safe_tags(estimator, key="allow_nan")}
+    tags = super(BaseForest, self).__sklearn_tags__()
+    tags.input_tags.allow_nan = get_tags(estimator).input_tags.allow_nan
+    return tags
 
 
-BaseForest._more_tags = _more_tags_patch
+BaseForest.__sklearn_tags__ = _sklearn_tags_patch
 
 
 class _BaseSurvivalForest(BaseForest, metaclass=ABCMeta):
@@ -104,7 +106,7 @@ def fit(self, X, y, sample_weight=None):
         """
         self._validate_params()
 
-        X = self._validate_data(X, dtype=DTYPE, accept_sparse="csc", ensure_min_samples=2, force_all_finite=False)
+        X = validate_data(self, X, dtype=DTYPE, accept_sparse="csc", ensure_min_samples=2, ensure_all_finite=False)
         event, time = check_array_survival(X, y)
 
         # _compute_missing_values_in_feature_mask checks if X has missing values and
@@ -115,7 +117,7 @@ def fit(self, X, y, sample_weight=None):
             X, estimator_name=self.__class__.__name__
         )
 
-        self.n_features_in_ = X.shape[1]
+        self._n_samples, self.n_features_in_ = X.shape
         time = time.astype(np.float64)
         self.unique_times_, self.is_event_time_ = get_unique_times(time, event)
         self.n_outputs_ = self.unique_times_.shape[0]
@@ -125,7 +127,18 @@ def fit(self, X, y, sample_weight=None):
         y_numeric[:, 1] = event.astype(np.float64)
 
         # Get bootstrap sample size
-        n_samples_bootstrap = _get_n_samples_bootstrap(n_samples=X.shape[0], max_samples=self.max_samples)
+        if not self.bootstrap and self.max_samples is not None:  # pylint: disable=no-else-raise
+            raise ValueError(
+                "`max_sample` cannot be set if `bootstrap=False`. "
+                "Either switch to `bootstrap=True` or set "
+                "`max_sample=None`."
+            )
+        elif self.bootstrap:
+            n_samples_bootstrap = _get_n_samples_bootstrap(n_samples=X.shape[0], max_samples=self.max_samples)
+        else:
+            n_samples_bootstrap = None
+
+        self._n_samples_bootstrap = n_samples_bootstrap
 
         # Check parameters
         self._validate_estimator()
@@ -141,13 +154,12 @@ def fit(self, X, y, sample_weight=None):
 
         n_more_estimators = self.n_estimators - len(self.estimators_)
 
-        if n_more_estimators < 0:
+        if n_more_estimators < 0:  # pylint: disable=no-else-raise
             raise ValueError(
                 f"n_estimators={self.n_estimators} must be larger or equal to "
                 f"len(estimators_)={len(self.estimators_)} when warm_start==True"
             )
-
-        if n_more_estimators == 0:
+        elif n_more_estimators == 0:
             warnings.warn("Warm-start fitting without increasing n_estimators does not fit new trees.", stacklevel=2)
         else:
             if self.warm_start and len(self.estimators_) > 0:
@@ -442,7 +454,7 @@ class RandomSurvivalForest(SurvivalAnalysisMixin, _BaseSurvivalForest):
     `min_impurity_decrease` or `min_impurity_split` are absent.
     In addition, the `feature_importances_` attribute is not available.
     It is recommended to estimate feature importances via
-    `permutation-based methods <https://eli5.readthedocs.io>`_.
+    :func:`sklearn.inspection.permutation_importance`.
 
     The features are always randomly permuted at each split. Therefore,
     the best found split may vary, even with the same training data,

sksurv/functions.py

@@ -12,7 +12,7 @@
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
 import numpy as np
-from sklearn.utils import check_consistent_length
+from sklearn.utils.validation import check_consistent_length
 
 __all__ = ["StepFunction"]