sklearn支持多类别(Multiclass)分类和多标签(Multilabel)分类：多类别分类：超过两个类别的分类任务。多类别分类假设每个样本属于且仅属于一个标签，类如一个水果可以是苹果或者是桔子但是不能同时属于两者。

多标签分类：给每个样本分配一个或多个标签。例如一个新闻可以既属于体育类，也属于文娱类。

sklearn的官方文档给出了支持多标签分类的类，包括如下：

以决策树举例，给出如下实现过程

数据准备

from sklearn.datasets import make_multilabel_classification

# Generate a random multilabel classification problem.

# For each sample, the generative process is:

# pick the number of labels: n ~ Poisson(n_labels)

# n times, choose a class c: c ~ Multinomial(theta)

# pick the document length: k ~ Poisson(length)

k times, choose a word: w ~ Multinomial(theta_c)

X, Y = datasets.make_multilabel_classification(n_samples=10, n_features=5, n_classes=3, n_labels=2)

生成的X和Y为如下形式的数据：

分类

from sklearn.model_selection import train_test_split

from sklearn.tree import DecisionTreeClassifier

# Split dataset to 8:2

X_train, X_test, Y_train ,Y_test = train_test_split(X, Y, test_size=0.2)

cls = DecisionTreeClassifier()

cls.fit(X_train, Y_train)

多标签分类评估

from sklearn import metrics

Y_pred = cls.predict(X_test)

Y_test和Y_pred值如下：

metrics.f1_score(Y_test, Y_pred, average="macro")

# 0.666

metrics.f1_score(Y_test, Y_pred, average="micro")

# 0.8

metrics.f1_score(Y_test, Y_pred, average="weighted")

# 1.0

metrics.f1_score(Y_test, Y_pred, average="samples")

# 0.4

概率预测

Y_prob = cls.predict_proba(X_test)

X_test和Y_prob值如下：predict_proba(X)

X：array-like or sparse matrix of shape = [n_samples, n_features]

RETURN：array of shape = [n_samples, n_classes], or a list of n_outputs