主要就以下函数作出举例:
Ⅰ.preprocessing.MinMaxScaler(feature_range=(-1, 1)).fit(X)
Ⅱ.scaler1.inverse_transform(X_scaled1)
Ⅲ.preprocessing.StandardScaler().fit(X)
Ⅳ.scaler2.inverse_transform(X_scaled2)
scaler1,scaler2为归一化标准化后地对象。
import torch
from sklearn import preprocessing
import numpy as np
X = np.array([[ 1., -1., 2.],[ 2., 0., 0.],[ 0., 1., -1.]])
print(X)
# The transformation is given by::
# X_std = (X - X.min(axis=0)) / (X.max(axis=0) - X.min(axis=0))
# X_scaled = X_std * (max - min) + min
# where min, max = feature_range.
# The transformation is calculated as::
# X_scaled = scale * X + min - X.min(axis=0) * scale
# where scale = (max - min) / (X.max(axis=0) - X.min(axis=0))
# 归一化(任意范围)
scaler1= preprocessing.MinMaxScaler(feature_range=(-1, 1)).fit(X)#归一化到(-1,1),scaler1保存最大最小值,便于回到原来参数
print(scaler1)
X_scaled1 = scaler1.transform(X)#下次归一化,按scaler1归一化到(-1,1)
print("归一化X_scaled1:\n{}".format(X_scaled1))
X1=scaler1.inverse_transform(X_scaled1)#按scaler1回到原来参数
print(X1)
print("fuck"*30)
# 标准化(标准高斯分布)<概率论基础啊>
scaler2 = preprocessing.StandardScaler().fit(X)#标准化到,scaler2保存方差和均值
print(scaler2)
X_scaled2 = preprocessing.scale(X)#下次标准化,按scaler2标准化到(mean_=0,var_=1)
print("标准化X_scaled2:\n{}".format(X_scaled2))
X2=scaler2.inverse_transform(X_scaled2)#按scaler2回到原来参数
print(X2)
print("*"*100)
print("scaler2.var_:{}".format(scaler2.var_))
print("scaler2.mean_:{}".format(scaler2.mean_))
print("&"*100)
print("scaler1.data_min_:{}".format(scaler1.data_min_))
print("scaler1.data_max_:{}".format(scaler1.data_max_))
效果:
黑色背景可能更好点:
具体的回归预测例子:pre_data是数据集,293*14,第一列是y,其他13个特征x:
x_data=pre_data[:,1:]
y_data=pre_data[:,0]
#仅对输入x_data作标准化
scaler2 = preprocessing.StandardScaler().fit(x_data)#标准化到,scaler2保存方差和均值
standar_data = preprocessing.scale(x_data)#下次标准化,按scaler2标准化到(mean_=0,var_=1)
x_train,x_test,y_train,y_test =train_test_split( standar_data ,
y_data ,test_size=0.15)
initial_data = scaler2.inverse_transform(x_test)#按scaler2回到原来x_data
plt.figure(1, figsize=(32, 20))
plt.plot(pre_data[:,13],pre_data[:,0], 'b-', label = 'actual')
plt.plot(initial_data[:,12] ,predictions, 'r.', label = 'prediction')# predictions是模型预测值
plt.show()
喜欢请点赞收藏,欢迎指正,谢谢。