Data preprocessing of Python machine learning

#数据预处理方法, mainly dealing with the dimension of data and the problem of the same trend.

Import NumPy as NP

From Sklearn Import preprocessing

#零均值规范

Data=np.random.rand (3,4) #随机生成3行4列的数据

Data_standardized=preprocessing.scale (data) #对数据进行归一化处理, that is, each value minus the mean divided by the variance is primarily used for SVM

#线性数据变换最大最小化处理

Data_scaler=preprocessing. Minmaxscaler (feature_range= (0,1)) #选定区间 (0,1), raw Data-min/(max-min)

Data_scaled=data_scaler.fit (data)

#数据标准化处理normalized

data_normalized=preprocessing.normalize (data,norm= ' L1 ') #减少人为增加特征, processed data Jia equals 1

#特征二值化,

Data_binarized=prepressing. Binarizer (threshold=0.5). Transform (data) #以0.5 is a threshold value greater than 0.5 is 1 and less than 0.5 is 0

#label_encode对标签进行数值化

Label_encode=preprocessing. Labelencoder ()

input_class=[' Audi ', ' Ford ', ' Audi ', ' BMW ', ' Toyota ', ' Benz '

Label_encode.fit (Input_class)

For I, item in Enmerate (LABEL_ENCODE.CLASS_):

Print (item, '--', i)

#onehotencode

Data preprocessing of Python machine learning