change ValueError to RuntimeWarning when n_dim < n_informative

.circleci/config.yml

@@ -66,7 +66,7 @@ jobs:
     docker:
       # CircleCI maintains a library of pre-built images
       # documented at https://circleci.com/doc/2.0/circleci-images/
-      - image: cimg/python:3.9
+      - image: cimg/python:3.11
     steps:
       - checkout
       - check-skip
@@ -111,7 +111,7 @@ jobs:
     docker:
       # CircleCI maintains a library of pre-built images
       # documented at https://circleci.com/doc/2.0/circleci-images/
-      - image: cimg/python:3.9
+      - image: cimg/python:3.11
     steps:
       - attach_workspace:
           at: ~/
@@ -128,7 +128,7 @@ jobs:
             ls build-install
             echo "After..."
             ls build-install/**
-            export PYTHONPATH=$PWD/build-install/usr/lib/python3.8/site-packages
+            export PYTHONPATH=$PWD/build-install/usr/lib/python3.11/site-packages
             ./spin docs
 
       - store_artifacts:

build_requirements.txt

@@ -3,7 +3,7 @@ meson-python>=0.16.0
 cython>=3.0.10
 ninja
 numpy
-scikit-learn>=1.4.2
+scikit-learn>=1.5.0
 click
 rich-click
 doit

doc/whats_new/v0.8.rst

@@ -18,6 +18,9 @@ Changelog
     encoded as infinity value. This is now fixed, and the estimators will raise an
     ValueError if missing-values are encountered in ``X`` input array.
     By `Adam Li`_ (:pr:`#264`)
+- |Feature| Simulations in ``sktree.datasets.hyppo`` now throw a warning instead
+    of an error when the number of samples is less than the number of dimensions.
+    By `Sambit Panda`_ (:pr:`#279`)
 
 Code and Documentation Contributors
 -----------------------------------

sktree/datasets/hyppo.py

@@ -1,3 +1,5 @@
+import warnings
+
 import numpy as np
 from scipy.integrate import nquad
 from scipy.stats import entropy, multivariate_normal
@@ -159,9 +161,11 @@ class distribution approaches the first class distribution by a factor of :math:
     .. footbibliography::
     """
     if n_dim < n_informative:
-        raise ValueError(
-            f"Number of informative dimensions {n_informative} must be less than number "
-            f"of dimensions, {n_dim}"
+        n_informative = n_dim
+        warnings.warn(
+            "Number of informative dimensions {n_informative} must be less than number "
+            f"of dimensions, {n_dim}. Setting n_informative to n_dim.",
+            RuntimeWarning,
         )
     if simulation not in MARRON_WAND_SIMS.keys():
         raise ValueError(
@@ -347,9 +351,11 @@ def make_trunk_mixture_classification(
     .. footbibliography::
     """
     if n_dim < n_informative:
-        raise ValueError(
-            f"Number of informative dimensions {n_informative} must be less than number "
-            f"of dimensions, {n_dim}"
+        n_informative = n_dim
+        warnings.warn(
+            "Number of informative dimensions {n_informative} must be less than number "
+            f"of dimensions, {n_dim}. Setting n_informative to n_dim.",
+            RuntimeWarning,
         )
     if mix < 0 or mix > 1:  # type: ignore
         raise ValueError("Mix must be between 0 and 1.")
@@ -506,9 +512,11 @@ def make_trunk_classification(
     .. footbibliography::
     """
     if n_dim < n_informative:
-        raise ValueError(
-            f"Number of informative dimensions {n_informative} must be less than number "
-            f"of dimensions, {n_dim}"
+        n_informative = n_dim
+        warnings.warn(
+            "Number of informative dimensions {n_informative} must be less than number "
+            f"of dimensions, {n_dim}. Setting n_informative to n_dim.",
+            RuntimeWarning,
         )
     rng = np.random.default_rng(seed=seed)
 

sktree/datasets/tests/test_hyppo.py

@@ -90,7 +90,7 @@ def test_make_trunk_generator_errors(trunk_gen):
         trunk_gen(n_samples=50, rho=0.5, band_type="invalid_band_type")
 
     # Test with an invalid band type
-    with pytest.raises(ValueError, match="Number of informative dimensions"):
+    with pytest.warns(RuntimeWarning, match="Number of informative dimensions"):
         trunk_gen(n_samples=50, n_dim=10, n_informative=11, rho=0.5)
 
 

sktree/experimental/monte_carlo.py

@@ -4,7 +4,8 @@
 from numpy.typing import ArrayLike
 from scipy.sparse import issparse
 from sklearn.neighbors import NearestNeighbors
-from sklearn.utils import _approximate_mode, _safe_indexing, check_array, check_consistent_length
+from sklearn.utils import _safe_indexing, check_array, check_consistent_length
+from sklearn.utils.extmath import _approximate_mode
 
 
 def _conditional_shuffle(nbrs: ArrayLike, replace: bool = False, seed=None) -> ArrayLike: