Python 探索性数据分析方法与应用

代码

# -*- coding: utf-8 -*-
#频率 and 众数
freDict {}
#统计元素数量
def count(l):
 for item in l:
 if item in freDict.keys():
 freDict[item] 1
 else:
 freDict[item] 1
 return
#求元素频率
def transform():
 s float(sum(freDict.values()))
 for k in freDict.keys():
 freDict[k] / s
 return
#获取众数
def mode():
 max_value list(freDict.values())[0]
 max_key list(freDict.keys())[0]
 for key,value in freDict.items():
 if value max_value:
 max_value value
 max_key key
 return max_key
l [ a , b , c , d , e , a , b , c , d , e , c , d , e ]
count(l)
print( Count for Distinct: ,freDict)
transform()
print( Frequency by precent: ,freDict)
print( Mode: ,mode())

输出结果

代码

# -*- coding: utf-8 -*-
from math import *
import numpy as np
iris np.loadtxt(r E:iris_proc.data ,delimiter , )
rel np.linspace(0,0, 11*5).reshape(11,5)
rel[...,0] range(0,101,10)
for col in range(1,5):
 rel[...,col] [np.percentile(iris[...,col-1], p) for p in rel[...,0]]
print(rel)

输出结果

代码

# -*- coding: utf-8 -*-
import math
import numpy as np
def mean(x):
 return sum(x)/ float(len(x))
def median(y):
 x np.sort(y)
 if len(x)%2 0:
 return (x[len(x)//2] x[len(x)//2 1])/2.0
 else:
 return x[len(x)//2]
def trimmean(x,p):
 b np.percentile(x, p//2)
 t np.percentile(x, 100-p//2)
 return mean([i for i in x if b i t])