-
Notifications
You must be signed in to change notification settings - Fork 0
/
PearsonSelection.py
32 lines (25 loc) · 1.18 KB
/
PearsonSelection.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
# from scipy.stats import pearsonr
def feature_selection(data):
data = np.array(data,dtype=float).reshape(-1,9)
data = pd.DataFrame(data)
# data.drop([data.columns[[1, 6, 8]]], axis=1, inplace=True)
data.drop([1,6,8],axis=1,inplace=True)
# drop columns "group","VEC", "AN"
data = np.array(data,dtype=float)
return data
if __name__ == '__main__':
data = pd.read_csv('../data/my_data.csv')
# for col_name, c_data in new_data.iteritems():
data = np.array(data)
Adata = pd.DataFrame(data.reshape(-1,9),columns=['period' ,'group' ,r'$r$' ,'CN' ,'At site' ,r'$\chi$', 'VEC' ,'M','1/AN'])
plt.figure()
corr_values = Adata.corr() # pandas直接调用corr就能计算特征之间的相关系数
sns.heatmap(corr_values, annot=True, vmax=1, square=True, cmap="Blues",fmt='.2f')
plt.tight_layout()
# plt.savefig('..\result\Pearson_value.pdf', format='pdf',bbox_inches='tight', dpi=1200)
plt.savefig(r'..\result\Pearson_value.svg', bbox_inches='tight', dpi=1200)
plt.show()