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 33 34 35
|
#!/usr/bin/python
# -*- coding: utf-8 -*-
##### importation des bibliothéques #####
import pickle
import numpy as np
import matplotlib.pyplot as plt
from scipy import stats
from utils.struct import filter_data
import seaborn as sns
sns.set(context="paper",font="monospace")
import pdb
from sklearn.feature_selection import SelectKBest, chi2, f_regression
##### importation de données #####
data=load(open("test.xlsx","rb"))
mesures_names_tri = ['var1', 'var2','var3','var4' ]
####### rejet de valeur extrémes
print len(data['var1'])
angles=np.unique(data['var4'])
#for mesure_name in mesures_names_tri:
mesure1=np.empty(1,float)
for ang in angles:
i_angle= [data['var4'] == ang]
mesure=data['var1'][i_angle]
k= stats.scoreatpercentile(mesure,99)
l= stats.scoreatpercentile(mesure,1)
for i in range(len(data['var1'][i_angle])):
mesure= np.delete(mesure,np.where(mesure < l) ,0)
mesure= np.delete(mesure,np.where(mesure > k) ,0)
mesure1=np.append(mesure1,mesure)
print (len(mesure1)) |
Partager