Deal with changing behavior of check_is_fitted

This is going to change in version 0.23, so we
might as well inline it now.

3 files changed, 20 insertions, 2 deletions

diff --git a/gensvm/core.py b/gensvm/core.py
index bfd5d9a..45d59ad 100644
--- a/gensvm/core.py
+++ b/gensvm/core.py
@@ -16,9 +16,9 @@ from sklearn.exceptions import ConvergenceWarning, FitFailedWarning
 from sklearn.preprocessing import LabelEncoder
 from sklearn.utils import check_array, check_X_y, check_random_state
 from sklearn.utils.multiclass import type_of_target
-from sklearn.utils.validation import check_is_fitted
 
 from .cython_wrapper import wrapper
+from .util import check_is_fitted
 
 
 def _fit_gensvm(
diff --git a/gensvm/sklearn_util.py b/gensvm/sklearn_util.py
index 182f257..e23921b 100644
--- a/gensvm/sklearn_util.py
+++ b/gensvm/sklearn_util.py
@@ -207,7 +207,12 @@ def _skl_check_is_fitted(estimator, method_name, refit):
             "attribute" % (type(estimator).__name__, method_name)
         )
     else:
-        check_is_fitted(estimator, "best_estimator_")
+        if not hasattr(estimator, "best_estimator_"):
+            raise NotFittedError(
+                "This %s instance is not fitted yet. Call "
+                "'fit' with appropriate arguments before using this "
+                "estimator." % type(estimator).__name__
+            )
 
 
 def _skl_grid_score(X, y, scorer_, best_estimator_, refit, multimetric_):
diff --git a/gensvm/util.py b/gensvm/util.py
index 046f3be..40d0eb1 100644
--- a/gensvm/util.py
+++ b/gensvm/util.py
@@ -8,6 +8,7 @@ Utility functions for GenSVM
 
 import numpy as np
 
+from sklearn.exceptions import NotFittedError
 
 def get_ranks(a):
     """
@@ -37,3 +38,15 @@ def get_ranks(a):
     ranks[~np.isnan(orig)] = count[dense - 1] + 1
     ranks[np.isnan(orig)] = np.max(ranks) + 1
     return list(ranks)
+
+
+def check_is_fitted(estimator, attribute):
+    msg = (
+        "This %(name)s instance is not fitted yet. Call 'fit' "
+        "with appropriate arguments before using this estimator."
+    )
+    if not hasattr(estimator, "fit"):
+        raise TypeError("%s is not an estimator instance" % (estimator))
+
+    if not hasattr(estimator, attribute):
+        raise NotFittedError(msg % {"name": type(estimator).__name__})