import pandas as pd
<<<<<<< HEAD
import numpy as np
from  scipy.stats import zscore
from sklearn.decomposition import PCA
=======
from  scipy.stats import zscore
>>>>>>> remotes/origin/盘荣博
import matplotlib.pyplot as plt
from matplotlib.pyplot import ylabel
df = pd.read_excel("棉花产量论文作业的数据.xlsx")
# plt.plot(df["年份"],df["单产"])
plt.rcParams['font.sans-serif']="SimHei"
# plt.rcParams['size'] =10
# plt.ylabel('单产')
# plt.xlabel('年份')

# print(df)
d = df.to_numpy()[:,1:]
print(d)
plt.subplot(4,1,1)
plt.scatter(d[:,:1],d[:,1:2],c='r')
ylabel('原始数据'),plt.title("单产和种子费用的关系")
#公式调用标准化，遵守标准正态分布
data = zscore(d)
print(data)
plt.subplot(4,1,2)
plt.scatter(data[:,:1],data[:,1:2],c='b',)
ylabel('zscore')

print(d.max(axis=0))
print(d.std(axis=0))
print(d.mean(axis=0))
#手写标准正态分布
data1=(d-d.mean(axis=0))/d.std(axis=0)
print(data1)
plt.subplot(4,1,3)
plt.scatter(data1[:,:1],data1[:,1:2],c='y')
ylabel('手写标准正态分布')

data2=(d-d.min(axis=0))/(d.max(axis=0)-d.min(axis=0))
plt.subplot(4,1,4)
plt.scatter(data2[:,:1],data2[:,1:2],c='g')
plt.xlabel('压缩到0~1')
print(data==data1)

<<<<<<< HEAD
# plt.savefig("shuju.jpg",dpi=2000)
# plt.show()
md= PCA().fit(data)
cf = np.cov(data.T)#求协方差矩阵
print(cf)
c, d= np.linalg.eig(cf)
print("特征值：\n",c)
print(md.explained_variance_)
e=c/c.sum()
# for _ in range(len(e)):
#     if(_!=0):
#         e[_]+=e[_-1]
print('贡献率：')
print(e)
print(md.explained_variance_ratio_)
print('特征向量：')
print(d.T)
print(md.components_)
print(md.components_-d.T<=0.1)
=======
plt.savefig("shuju.jpg",dpi=2000)
plt.show()
>>>>>>> remotes/origin/盘荣博