scikit-survival 0.25.0__cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl

sksurv/testing.py

@@ -0,0 +1,108 @@
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+from importlib import import_module
+import inspect
+from pathlib import Path
+import pkgutil
+
+import numpy as np
+from numpy.testing import assert_almost_equal, assert_array_equal
+import pytest
+from sklearn.base import BaseEstimator, TransformerMixin
+
+import sksurv
+from sksurv.metrics import concordance_index_censored
+
+
+def assert_cindex_almost_equal(event_indicator, event_time, estimate, expected):
+    result = concordance_index_censored(event_indicator, event_time, estimate)
+    assert_array_equal(result[1:], expected[1:])
+    concordant, discordant, tied_risk = result[1:4]
+    cc = (concordant + 0.5 * tied_risk) / (concordant + discordant + tied_risk)
+    assert_almost_equal(result[0], cc)
+    assert_almost_equal(result[0], expected[0])
+
+
+def assert_survival_function_properties(surv_fns):
+    if not np.isfinite(surv_fns).all():
+        raise AssertionError("survival function contains values that are not finite")
+    if np.any(surv_fns < 0.0):
+        raise AssertionError("survival function contains negative values")
+    if np.any(surv_fns > 1.0):
+        raise AssertionError("survival function contains values larger 1")
+
+    d = np.apply_along_axis(np.diff, 1, surv_fns)
+    if np.any(d > 0):
+        raise AssertionError("survival functions are not monotonically decreasing")
+
+    # survival function at first time point
+    num_closer_to_zero = np.sum(1.0 - surv_fns[:, 0] >= surv_fns[:, 0])
+    if num_closer_to_zero / surv_fns.shape[0] > 0.5:
+        raise AssertionError(f"most ({num_closer_to_zero}) probabilities at first time point are closer to 0 than 1")
+
+    # survival function at last time point
+    num_closer_to_one = np.sum(1.0 - surv_fns[:, -1] < surv_fns[:, -1])
+    if num_closer_to_one / surv_fns.shape[0] > 0.5:
+        raise AssertionError(f"most ({num_closer_to_one}) probabilities at last time point are closer to 1 than 0")
+
+
+def assert_chf_properties(chf):
+    if not np.isfinite(chf).all():
+        raise AssertionError("chf contains values that are not finite")
+    if np.any(chf < 0.0):
+        raise AssertionError("chf contains negative values")
+
+    d = np.apply_along_axis(np.diff, 1, chf)
+    if np.any(d < 0):
+        raise AssertionError("chf are not monotonically increasing")
+
+    # chf at first time point
+    num_closer_to_one = np.sum(1.0 - chf[:, 0] < chf[:, 0])
+    if num_closer_to_one / chf.shape[0] > 0.5:
+        raise AssertionError(f"most ({num_closer_to_one}) hazard rates at first time point are closer to 1 than 0")
+
+
+def _is_survival_estimator(x):
+    return (
+        inspect.isclass(x)
+        and issubclass(x, BaseEstimator)
+        and not issubclass(x, TransformerMixin)
+        and x.__module__.startswith("sksurv.")
+        and not x.__name__.startswith("_")
+        and x.__module__.split(".", 2)[1] not in {"metrics", "nonparametric"}
+    )
+
+
+def all_survival_estimators():
+    root = str(Path(sksurv.__file__).parent)
+    all_classes = []
+    for _importer, modname, _ispkg in pkgutil.walk_packages(path=[root], prefix="sksurv."):
+        # meta-estimators require base estimators
+        if modname.startswith("sksurv.meta"):
+            continue
+        module = import_module(modname)
+        for _name, cls in inspect.getmembers(module, _is_survival_estimator):
+            if inspect.isabstract(cls):
+                continue
+            all_classes.append(cls)
+    return set(all_classes)
+
+
+class FixtureParameterFactory:
+    def get_cases(self):
+        cases = []
+        for name, func in inspect.getmembers(self):
+            if name.startswith("data_"):
+                values = func()
+                cases.append(pytest.param(*values, id=name))
+        return cases

sksurv/tree/__init__.py

@@ -0,0 +1 @@
+from .tree import ExtraSurvivalTree, SurvivalTree  # noqa: F401