本文整理汇总了Python中sklearn.tree.DecisionTreeClassifier.transform方法的典型用法代码示例。如果您正苦于以下问题:Python DecisionTreeClassifier.transform方法的具体用法?Python DecisionTreeClassifier.transform怎么用?Python DecisionTreeClassifier.transform使用的例子?那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类sklearn.tree.DecisionTreeClassifier
的用法示例。
在下文中一共展示了DecisionTreeClassifier.transform方法的1个代码示例,这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞,您的评价将有助于系统推荐出更棒的Python代码示例。
示例1: Customer
# 需要导入模块: from sklearn.tree import DecisionTreeClassifier [as 别名]
# 或者: from sklearn.tree.DecisionTreeClassifier import transform [as 别名]
#Mean Feature Importance
print "Mean Feature Importance %.6f" %np.mean(importances)
# In[9]:
#Okay so looks like some features are much more important than the others. Attribute 0 for instance is the status of
#checking accounts of the Customer (Importance 13%) while Attribute 7 is how long they've been employed with the
#current employer (3.6%). It makes sense that one matters more than other.
#Let's do some trimming. We'll try to transform the Training set to include only features that are atleast as important as
#the mean of importances. Let's find out if this improves the accuracy.
#Luckily this is very easy to do in sklearn. RF has a transform method that helps with this.
X_train_r = estimator.transform(X_train, threshold='mean')
X_test_r = estimator.transform(X_test, threshold='mean')
#Let's run the learning curve again.
title = "Learning Curves -Iter2 (Random Forests, n_estimators=%.6f, feature_importances > mean)" %(n_estimators)
estimator = RandomForestClassifier(n_estimators=n_estimators, max_depth=max_depth, n_jobs=10)
plot_learning_curve(estimator, title, X_train_r, y_train, cv=cv)
plt.show()
#Looks like that didn't really have much of an impact on the model. Let's find out how well the model will generalize
#by predicting on the Test dataset.
# In[11]:
#Let's call fit on the estimator so we can look at feature importances.