Error during execution

on implementing this on colab after git clone:
!python /content/probability/tensorflow_probability/examples/bayesian_neural_network.py

`Traceback (most recent call last):
File "/usr/local/lib/python3.6/dist-packages/seaborn/_core.py", line 1456, in categorical_order
order = vector.cat.categories
AttributeError: 'tensorflow.python.framework.ops.EagerTensor' object has no attribute 'cat'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "/usr/local/lib/python3.6/dist-packages/seaborn/_core.py", line 1460, in categorical_order
order = vector.unique()
AttributeError: 'tensorflow.python.framework.ops.EagerTensor' object has no attribute 'unique'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "/content/probability/tensorflow_probability/examples/bayesian_neural_network.py", line 366, in
app.run(main)
File "/usr/local/lib/python3.6/dist-packages/absl/app.py", line 300, in run
_run_main(main, args)
File "/usr/local/lib/python3.6/dist-packages/absl/app.py", line 251, in _run_main
sys.exit(main(argv))
File "/content/probability/tensorflow_probability/examples/bayesian_neural_network.py", line 362, in main
.format(heldout_log_prob))
File "/content/probability/tensorflow_probability/examples/bayesian_neural_network.py", line 150, in plot_heldout_prediction
sns.barplot(np.arange(10), prob_sample[i, :], alpha=0.1, ax=ax)
File "/usr/local/lib/python3.6/dist-packages/seaborn/_decorators.py", line 46, in inner_f
return f(**kwargs)
File "/usr/local/lib/python3.6/dist-packages/seaborn/categorical.py", line 3172, in barplot
errcolor, errwidth, capsize, dodge)
File "/usr/local/lib/python3.6/dist-packages/seaborn/categorical.py", line 1585, in init
order, hue_order, units)
File "/usr/local/lib/python3.6/dist-packages/seaborn/categorical.py", line 203, in establish_variables
group_names = categorical_order(groups, order)
File "/usr/local/lib/python3.6/dist-packages/seaborn/_core.py", line 1462, in categorical_order
order = pd.unique(vector)
File "/usr/local/lib/python3.6/dist-packages/pandas/core/algorithms.py", line 390, in unique
uniques = _reconstruct_data(uniques, original.dtype, original)
File "/usr/local/lib/python3.6/dist-packages/pandas/core/algorithms.py", line 208, in _reconstruct_data
values = values.astype(dtype, copy=False)
TypeError: data type not understood`

Start with tensorflow_probability/examples/bayesian_neural_network.py, especially plot_heldout_prediction around line 150, and reproduce the Colab command from the report. Compare the plotting inputs with the seaborn and pandas traceback; done means the example runs past this plotting step without the reported error.