[BUG] `np.<ufunc>()` returns `None` when passed a `cudf.MultiIndex`
Open
0 - Backlog
bug
Python
- Dominant language
- C++
- Stars
- 9.8k
- Forks
- 1.1k
- Avg merge
- 3d 6m
- Merged PRs (30d)
- 278
Description
**Describe the bug**
`np.exp` silently returns `None` for `cudf.MultiIndex`
**Steps/Code to reproduce bug**
```
In [1]: import numpy as np
In [2]: import cudf
In [3]: np.exp(cudf.MultiIndex.from_tuples([(1, 2), (3, 4)])) # returns None
```
**Expected behavior**
This code should raise an error, which is what happens with a `pandas.MultiIndex`
```
In [4]: pandas as pd
In [5]: np.exp(pd.MultiIndex.from_tuples([(1, 2), (3, 4)]))
...
TypeError: loop of ufunc does not support argument 0 of type tuple which has no callable exp method
```
Contributor guide
Assessment
This issue has not been assessed yet.