TypeError when calling add_edges with columns of type object

[AZeed18]

I have a BNClassifier class that inherits from HybridBN and ClassifierMixin

Its fit method includes add_edges method from HybridBN

I get the following error (with Hill Climbing only):

  File "<HIDE>", line 59, in fit
    self.add_edges(df, scoring_function=('BIC',), params={'init_edges': <HIDE>})
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "<HIDE>/lib/python3.11/site-packages/bamt/networks/base.py", line 193, in add_edges
    worker.build(
  File "<HIDE>/lib/python3.11/site-packages/bamt/builders/hc_builder.py", line 198, in build
    self.apply_group1(data=data, progress_bar=progress_bar, **self.params)
  File "<HIDE>/lib/python3.11/site-packages/bamt/builders/hc_builder.py", line 121, in apply_group1
    bn = hc_method(
         ^^^^^^^^^^
  File "<HIDE>/lib/python3.11/site-packages/bamt/redef_HC.py", line 165, in hc
    cache[old_cols] = mutual_information(data[:, old_cols])
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "<HIDE>/lib/python3.11/site-packages/bamt/redef_info_scores.py", line 130, in BIC_local
    log_score = log_lik_local(data, method=method)
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "<HIDE>/lib/python3.11/site-packages/bamt/redef_info_scores.py", line 123, in log_lik_local
    mutual_information(data, method=method)
  File "<HIDE>/lib/python3.11/site-packages/bamt/mi_entropy_gauss.py", line 224, in mi_gauss
    return mi_gauss(loc_to_DataFrame(data), method, conditional)
                    ^^^^^^^^^^^^^^^^^^^^^^
  File "<HIDE>/lib/python3.11/site-packages/bamt/preprocess/numpy_pandas.py", line 16, in loc_to_DataFrame
    nodes_type = get_type_numpy(data)
                 ^^^^^^^^^^^^^^^^^^^^
  File "<HIDE>/lib/python3.11/site-packages/bamt/preprocess/numpy_pandas.py", line 50, in get_type_numpy
    all_row_is_integer = all(
                         ^^^^
  File "<HIDE>/lib/python3.11/site-packages/bamt/preprocess/numpy_pandas.py", line 51, in <genexpr>
    np.issubdtype(x, np.integer) or x.is_integer() for x in row
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "<HIDE>/lib/python3.11/site-packages/numpy/core/numerictypes.py", line 417, in issubdtype
    arg1 = dtype(arg1).type
           ^^^^^^^^^^^
TypeError: Cannot interpret '39' as a data type

39 is the value of the second column (an np.dtypes.Int32DType column) in the first row

Datatypes in my DataFrame are:

• np.dtypes.Float32DType for continuous data
• np.dtypes.Int32DType for discrete numerical data
• np.dtypes.ObjectDType for categorical data