diff --git a/tests/xgboost/test_xgboost_converters_base_score.py b/tests/xgboost/test_xgboost_converters_base_score.py
index 7f221855..7ce1697a 100644
--- a/tests/xgboost/test_xgboost_converters_base_score.py
+++ b/tests/xgboost/test_xgboost_converters_base_score.py
@@ -35,9 +35,8 @@ def test_xgbregressor_sparse_base_score(self):
         rf = XGBRegressor(n_estimators=3, max_depth=4, random_state=0, base_score=0.5)
         rf.fit(X_sp, y)
         expected = rf.predict(X).astype(np.float32).reshape((-1, 1))
-        expected_sparse = rf.predict(X_sp).astype(np.float32).reshape((-1, 1))
-        diff = np.abs(expected - expected_sparse)
-        self.assertNotEqual(diff.min(), diff.max(), atol=1e-5)
+        # expected sparse is expected ot be diffrent than expected,
+        # expected_sparse = rf.predict(X_sp).astype(np.float32).reshape((-1, 1))
 
         onx = convert_xgboost(
             rf,
@@ -64,9 +63,8 @@ def test_xgbregressor_sparse_no_base_score(self):
         rf = XGBRegressor(n_estimators=3, max_depth=4, random_state=0)
         rf.fit(X_sp, y)
         expected = rf.predict(X).astype(np.float32).reshape((-1, 1))
-        expected_sparse = rf.predict(X_sp).astype(np.float32).reshape((-1, 1))
-        diff = np.abs(expected - expected_sparse)
-        self.assertNotEqual(diff.min(), diff.max(), atol=1e-5)
+        # expected sparse is expected ot be diffrent than expected,
+        # expected_sparse = rf.predict(X_sp).astype(np.float32).reshape((-1, 1))
 
         onx = convert_xgboost(
             rf,
@@ -94,9 +92,8 @@ def test_xgbclassifier_sparse_base_score(self):
         rf = XGBClassifier(n_estimators=3, max_depth=4, random_state=0, base_score=0.5)
         rf.fit(X_sp, y)
         expected = rf.predict_proba(X).astype(np.float32).reshape((-1, 1))
-        expected_sparse = rf.predict_proba(X_sp).astype(np.float32).reshape((-1, 1))
-        diff = np.abs(expected - expected_sparse)
-        self.assertNotEqual(diff.min(), diff.max(), atol=1e-5)
+        # expected sparse is expected ot be diffrent than expected,
+        # expected_sparse = rf.predict_proba(X_sp).astype(np.float32).reshape((-1, 1))
 
         onx = convert_xgboost(
             rf,
@@ -124,9 +121,8 @@ def test_xgbclassifier_sparse_no_base_score(self):
         rf = XGBClassifier(n_estimators=3, max_depth=4, random_state=0)
         rf.fit(X_sp, y)
         expected = rf.predict_proba(X).astype(np.float32).reshape((-1, 1))
-        expected_sparse = rf.predict_proba(X_sp).astype(np.float32).reshape((-1, 1))
-        diff = np.abs(expected - expected_sparse)
-        self.assertNotEqual(diff.min(), diff.max(), atol=1e-5)
+        # expected sparse is expected ot be diffrent than expected,
+        # expected_sparse = rf.predict_proba(X_sp).astype(np.float32).reshape((-1, 1))
 
         onx = convert_xgboost(
             rf,