NO.58——100天机器学习实践第一天：数据预处理

这是github上的一个给力的项目，地址在这：https://github.com/MLEveryday/100-Days-Of-ML-Code

---

1.到入库

import numpy as np
import pandas as py

 

2.导入数据集 

dataset = pd.read_csv('Data.csv')
dataset

X = dataset.iloc[:,:-1].values   # [行，列] 除最后一列元素取全部
X
Y = dataset.iloc[:,3].values   #取下标为3的列
Y

3.处理丢失数据

from sklearn.impute import SimpleImputer
imputer = SimpleImputer(missing_values=np.nan, strategy="mean")  #用特征列的均值替代
imputer = imputer.fit(X[:, 1:3])  #左闭右开，取下标为1，2列
X[:, 1:3] = imputer.transform(X[:, 1:3])

4.解析分类数据

#引入标签编码和独热编码
from sklearn.preprocessing import LabelEncoder, OneHotEncoder
labelencoder_X = LabelEncoder()#先用labelencoder将字符串编码为0,1,2，再用onrhotencoder编码
X[ : , 0] = labelencoder_X.fit_transform(X[ : , 0])

#对训练集进行独热编码，对测试集进行标签编码
onehotencoder = OneHotEncoder(categories = 'auto')
X = onehotencoder.fit_transform(X).toarray()
labelencoder_Y = LabelEncoder()
Y = labelencoder_Y.fit_transform(Y)

5.拆分数据集为训练集和测试集

from sklearn.model_selection import train_test_split
X_train, X_test, Y_train, Y_test = train_test_split( X , Y , test_size = 0.2, random_state = 0)

6.特征量化

from sklearn.preprocessing import StandardScaler
sc_X = StandardScaler()  #标准化
X_train = sc_X.fit_transform(X_train)
X_test = sc_X.transform(X_test)