I tried to plot the classifier tree with catboost in jupyter notebook.
However, it seems that it is not supported by Python 3.
I followed this link for plotting but got error.
---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
<ipython-input-140-117bb3853789> in <module>
12 model.plot_tree(
13 tree_idx=0,
---> 14 pool=pool # "pool" is required parameter for trees with one hot features
15 )
~\AppData\Local\Continuum\anaconda3\lib\site-packages\catboost\core.py in plot_tree(self, tree_idx, pool)
2386 for node_num in range(layer_size):
2387 if split_num >= 0:
-> 2388 node_label = splits[split_num].decode('utf-8').replace('bin=', 'value>', 1)
2389 color = 'black'
2390 shape = 'ellipse'
AttributeError: 'str' object has no attribute 'decode'
It seems that string in python 3 no longer has decode method.(HERE)
Any suggestion?
