BUG: Fixed strange behaviour of pd.DataFrame.drop() with inplace argu…

pandas/core/ops/methods.py

@@ -93,7 +93,8 @@ def _wrap_inplace_method(method):
 
         def f(self, other):
             result = method(self, other)
-
+            # Delete cacher
+            self._reset_cacher()
             # this makes sure that we are aligned like the input
             # we are updating inplace so we want to ignore is_copy
             self._update_inplace(

pandas/tests/frame/test_axis_select_reindex.py

@@ -1152,3 +1152,22 @@ def test_drop_non_empty_list(self, index, drop_labels):
         # GH 21494
         with pytest.raises(KeyError, match="not found in axis"):
             pd.DataFrame(index=index).drop(drop_labels)
+
+    @pytest.mark.parametrize(
+        "operation", ["__iadd__", "__isub__", "__imul__", "__ipow__"]
+    )
+    @pytest.mark.parametrize("inplace", [False, True])
+    def test_inplace_drop_and_operation(self, operation, inplace):
+        # GH 30484
+        data_dict = {"x": [1, 2, 3, 4, 5], "y": [10, 20, 30, 40, 50]}
+        df = pd.DataFrame(data_dict)
+        y = df["y"]
+
+        if inplace:
+            df.drop("y", axis=1, inplace=inplace)
+        else:
+            df = df.drop("y", axis=1, inplace=inplace)
+        # Perform operation and ensure that df is not changed
+        expected = df.copy()
+        getattr(y, operation)(1)
+        tm.assert_frame_equal(df, expected)