Building the decision tree

As shown earlier, we can create a decision tree using scikit-learn's tree module. For now, let's not specify any optional arguments:

In [5]: from sklearn import tree...     dtc = tree.DecisionTreeClassifier()

Do you remember how to train the decision tree?

In [6]: dtc.fit(X_train, y_train)Out[6]: DecisionTreeClassifier(class_weight=None, criterion='gini',                               max_depth=None, max_features=None,                               max_leaf_nodes=None,                               min_impurity_split=1e-07,                               min_samples_leaf=1,                               min_samples_split=2,                               min_weight_fraction_leaf=0.0,                               presort=False, random_state=None,                               splitter='best')

Since we did not specify any pre-pruning parameters, we would expect this decision tree to grow quite large and result in a perfect score on the training set: ...

Get Machine Learning for OpenCV now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.