scikit-learn杂记

1.数据预处理 二值化

import numpy as np
from sklearn import preprocessing

X = np.array([[1., -1., 2.], [2., 0., 0.], [0., 1., -1.]])
binarized = preprocessing.Binarizer().fit(X)
print(binarized.transform(X))

2.数据预处理 Onehot处理离散数据

import numpy as np
from sklearn import preprocessing

Y = np.array([[0, 1, 0], [1, 0, 1], [2, 2, 1], [3, 1, 0]])
enc = preprocessing.OneHotEncoder()
enc.fit(Y)
print(enc.transform([[3, 0, 1]]).toarray())

3.综合处理文本离散数据 Onehot处理离散文本数据

import numpy as np
from sklearn import preprocessing
from sklearn.preprocessing import LabelEncoder

# 原始离散数据，其中国家有四种数据，职业有三种数据，性别有两种数据，即[2,3,4]
Y_label = np.array([['from China', 'Student', 'Male'], ['from USA', 'Teacher', 'Female'],
                    ['from UK', 'Engineer', 'Female'],['from AU', 'Student', 'Male']])

# 将离散文本转换为数字表示
le_from = LabelEncoder()
le_job = LabelEncoder()
le_gender = LabelEncoder()
le_from.fit(np.array(['from China', 'from USA', 'from UK', 'from AU']))
le_job.fit(np.array(['Student', 'Teacher', 'Engineer']))
le_gender.fit(np.array(['Male','Female']))

# 替换原数据
Y_label[:, 0] = le_from.transform(Y_label[:, 0])
Y_label[:, 1] = le_job.transform(Y_label[:, 1])
Y_label[:, 2] = le_gender.transform(Y_label[:, 2])

# 使用OneHot编码数据
enc = preprocessing.OneHotEncoder()
enc.fit(Y_label)
print(enc.transform([[3, 0, 1]]).toarray())