Fixed missing value to throw error on unify_data. Black reformat.

Author: interpret-ml

python/interpret-core/interpret/utils/all.py

@@ -298,10 +298,10 @@ def unify_data(data, labels=None, feature_names=None, feature_types=None):
 
     # NOTE: Until missing handling is introduced, all methods will fail at data unification stage if present.
     new_data_has_na = (
-        True if new_data is not None and np.isnan(new_data).any() else False
+        True if new_data is not None and pd.isnull(new_data).any() else False
     )
     new_labels_has_na = (
-        True if new_labels is not None and np.isnan(new_labels).any() else False
+        True if new_labels is not None and pd.isnull(new_labels).any() else False
     )
 
     if new_data_has_na or new_labels_has_na:

python/interpret-core/interpret/utils/environment.py

@@ -90,7 +90,15 @@ def _detect_databricks():
 
 
 def is_cloud_env(detected):
-    cloud_env = ["databricks", "azure", "azureml_vm", "kaggle", "sagemaker", "binder", "colab"]
+    cloud_env = [
+        "databricks",
+        "azure",
+        "azureml_vm",
+        "kaggle",
+        "sagemaker",
+        "binder",
+        "colab",
+    ]
     if len(set(cloud_env).intersection(detected)) != 0:
         return True
     else:

python/interpret-core/interpret/utils/test/test_utils.py

@@ -3,6 +3,7 @@
 
 import pytest
 import numpy as np
+import pandas as pd
 from .. import gen_feat_val_list, gen_name_from_class
 from .. import reverse_map, unify_data
 
@@ -20,6 +21,20 @@ def test_unify_fails_on_missing():
         unify_data(orig_data, orig_labels)
 
 
+def test_unify_dataframe_smoke():
+    df = pd.DataFrame()
+    df["f1"] = [1.5, "a"]
+    df["f2"] = [3, "b"]
+    df["label"] = [0, 1]
+
+    train_cols = df.columns[0:-1]
+    label = df.columns[-1]
+    X = df[train_cols]
+    y = df[label]
+
+    unify_data(X, y)
+
+
 def test_unify_list_data():
     orig_data = [[1, 2], [3, 4]]
     orig_labels = [0, 0]