Bag of Bonds featurizer

I have a pandas dataframe of structures from the QMOF database on the materials project. When I try to fit the bag of bonds featurizer:

featurizer = BagofBonds()['structure'][0:5])

I get the following:

ValueError                                Traceback (most recent call last)
Cell In[111], line 2
      1 featurizer = BagofBonds()
----> 2['structure'][0:5])

File ~/anaconda3/envs/py38/lib/python3.8/site-packages/matminer/featurizers/structure/, in, X, y)
    535 unpadded_bobs = [self.bag(s, return_baglens=True) for s in X]
    536 bonds = [list(bob.keys()) for bob in unpadded_bobs]
--> 537 bonds = np.unique(sum(bonds, []))
    538 baglens = [0] * len(bonds)
    540 for i, bond in enumerate(bonds):

File <__array_function__ internals>:200, in unique(*args, **kwargs)

File ~/anaconda3/envs/py38/lib/python3.8/site-packages/numpy/lib/, in unique(ar, return_index, return_inverse, return_counts, axis, equal_nan)
    138 @array_function_dispatch(_unique_dispatcher)
    139 def unique(ar, return_index=False, return_inverse=False,
    140            return_counts=False, axis=None, *, equal_nan=True):
    141     """
    142     Find the unique elements of an array.
    271     """
--> 272     ar = np.asanyarray(ar)
    273     if axis is None:
    274         ret = _unique1d(ar, return_index, return_inverse, return_counts, 
    275                         equal_nan=equal_nan)

ValueError: setting an array element with a sequence. The requested array has an inhomogeneous shape after 1 dimensions. The detected shape was (101,) + inhomogeneous part.

featurizer.bag(structure) works fine on the other hand. Can anyone help?

Hmm that is strange, could you send 2-3 example structures from your DF that this happens for? I will try running on my end and see if I can debug