Fix mutual_information bug with string index (#1199)

* fix mi bug with string index

* update release notes

Author: thehomebrewnerd

docs/source/release_notes.rst

@@ -7,14 +7,15 @@ Future Release
 ==============
     * Enhancements
     * Fixes
+        * Fix bug that causes ``mutual_information`` to fail with certain index types (:pr:`1199`)
     * Changes
         * Update pip to 21.3.1 for test requirements (:pr:`1196`)
     * Documentation Changes
         * Update install page with updated minimum optional dependencies (:pr:`1193`)
     * Testing Changes
 
     Thanks to the following people for contributing to this release:
-    :user:`gsheni`
+    :user:`gsheni`, :user:`thehomebrewnerd`
 
 v0.9.0 Nov 11, 2021
 ===================

woodwork/statistics_utils.py

@@ -280,8 +280,9 @@ def _get_mutual_information_dict(
         if type(col.logical_type) in valid_types
     ]
 
-    if not include_index and dataframe.ww.index is not None:
-        valid_columns.remove(dataframe.ww.index)
+    index = dataframe.ww.index
+    if not include_index and index is not None and index in valid_columns:
+        valid_columns.remove(index)
 
     data = dataframe.loc[:, valid_columns]
     if _is_dask_dataframe(data):

woodwork/tests/accessor/test_statistics.py

@@ -295,6 +295,23 @@ def test_get_valid_mi_columns_with_index(sample_df):
     assert "id" in mi
 
 
+def test_mutual_info_with_string_index():
+    df = pd.DataFrame(
+        {
+            "id": ["a", "b", "c"],
+            "col1": [1, 2, 3],
+            "col2": [10, 20, 30],
+        }
+    )
+    df.ww.init(index="id", logical_types={"id": "unknown"})
+    mi = df.ww.mutual_information()
+
+    cols_used = set(np.unique(mi[["column_1", "column_2"]].values))
+    assert "id" not in cols_used
+    assert "col1" in cols_used
+    assert "col2" in cols_used
+
+
 def test_get_describe_dict(describe_df):
     describe_df.ww.init(index="index_col")