统计函数demo

def to_array(x):
    # 公共函数转数据结构
    if isinstance(x,np.ndarray):
        return x
    else:
        if isinstance(x,(list,tuple)):
            return np.array(x)
        elif isinstance(x,set):
            return np.array(list(x))
        
to_array({1,2,3,4,5})

def mean(x):
    # 均值
    return sum(x)/len(x)

mean(counter)

def var(x):
    # 方差
    x = to_array(x)
    return (x - mean(x)).dot(x - mean(x))/(len(x)-1)
var([1,2,2,1,4])

def cov(x,y):
    # 协方差
    x,y = to_array(x),to_array(y)
    return (x-mean(x)).dot(y-mean(y))/(len(x)-1)

cov([1,2,3,4],[1,1,3,4])

# 标准差
def std(x):
    return np.sqrt(var(x))

std([1,2,3,4])

# 皮尔逊系数
def pearson(x,y):
    return cov(x,y)/(std(x)*std(y))

pearson([1,2,3,4],[1,3,2,4])

# 中值
def median(x):
    n = len(x)
    index = n // 2
    if (index*2)==n:
        return sum(sorted(x)[index-1:index+1])/2
    else:
        return sorted(x)[index]

median([1,2,3,4])

# 众数
def mode(x):
    dict_ = {}
    temp = 0
    for _ in x:
        if _ in dict_:
            dict_[_]+=1
        else:
            dict_.update({_:1})
        if dict_[_] > temp:
            temp = dict_[_] 
    expr = {}
    for k,v in dict_.items():
        if v == temp:
            expr.update({k:v})
  
    return expr,dict_
        
mode([0,1,1,2,2,2,2,1])

# 变异系数
def coe(x):
    return std(x)/mean(x)


coe([1,2,1,1,1,4,1,1])