统计相关
次序统计
计算最小值
- numpy.amin(a[, axis=None, out=None, keepdims=np._NoValue, initial=np._NoValue,where=np._NoValue])
Return the minimum of an array or minimum along an axis.
# 【例】计算最小值
import numpy as np
x = np.array(
[[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]])
y = np.amin(x)
print(y)
y = np.amin(x,axis=0)
print(y)
y = np.amin(x,axis=1)
print(y)
11
[11 12 13 14 15]
[11 16 21 26 31]
计算最大值
- numpy.amax(a[, axis=None, out=None, keepdims=np._NoValue, initial=np._NoValue,where=np._NoValue])
Return the maximum of an array or maximum along an axis.
# 【例】计算最大值
import numpy as np
x = np.array([[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]])
y = np.amax(x)
print(y) # 35
y = np.amax(x, axis=0)
print(y) # [31 32 33 34 35]
y = np.amax(x, axis=1)
print(y) # [15 20 25 30 35]
35
[31 32 33 34 35]
[15 20 25 30 35]
计算极差
- numpy.ptp(a, axis=None, out=None, keepdims=np._NoValue)
Range of values (maximum -minimum) along an axis. The name of the function comes from the acronym for ‘peak to
peak’.
import numpy as np
np.random.seed(20201127)
x = np.random.randint(0,20,size=[4,5])
print(x)
print(np.ptp(x))
print(np.ptp(x,axis=0))
print(np.ptp(x,axis=1))
[[15 13 2 2 16]
[ 5 5 13 10 16]
[ 1 19 5 12 0]
[17 8 7 14 2]]
19
[16 14 11 12 16]
[14 11 19 15]
计算分位数
- numpy.percentile(a, q, axis=None, out=None, overwrite_input=False,interpolation=‘linear’, keepdims=False)
Compute the q-th percentile of the data along the specified axis. Returns the q-th percentile(s) of the array elements.
-
a:array,用来算分位数的对象,可以是多维的数组。
-
q:介于0-100的float,用来计算是几分位的参数,如四分之一位就是25,如要算两个位置的数就[25,75]。
-
axis:坐标轴的方向,一维的就不用考虑了,多维的就用这个调整计算的维度方向,取值范围0/1。
import numpy as np
np.random.seed(20201127)
x = np.random.randint(0,20,size=[4,5])
print(x)
print(np.percentile(x,[25,50]))
print(np.percentile(x,[25,50]))
print(np.percentile(x,[25,50]))
print(np.percentile(x,[25,50],axis=0))
print(np.percentile(x,[20,50],axis=1))
[[15 13 2 2 16]
[ 5 5 13 10 16]
[ 1 19 5 12 0]
[17 8 7 14 2]]
[4.25 9. ]
[4.25 9. ]
[4.25 9. ]
[[ 4. 7.25 4.25 8. 1.5 ]
[10. 10.5 6. 11. 9. ]]
[[ 2. 5. 0.8 6. ]
[13. 10. 5. 8. ]]
均值与方差
计算中位数
- numpy.median(a, axis=None, out=None, overwrite_input=False, keepdims=False) Compute
the median along the specified axis. Returns the median of the array elements.
# 【例】计算中位数
import numpy as np
np.random.seed(20201127)
x = np.random.randint(0,20,size=[4,5])
print(x)
print(np.percentile(x,50))
print(np.median(x))
print(np.percentile(x,50,axis=0))
print(np.median(x,axis=0))
print(np.percentile(x,50,axis=1))
print(np.median(x,axis=1))
[[15 13 2 2 16]
[ 5 5 13 10 16]
[ 1 19 5 12 0]
[17 8 7 14 2]]
9.0
9.0
[10. 10.5 6. 11. 9. ]
[10. 10.5 6. 11. 9. ]
[13. 10. 5. 8.]
[13. 10. 5. 8.]
计算平均值
- numpy.mean(a[, axis=None, dtype=None, out=None, keepdims=np._NoValue)])
Compute the arithmetic mean along the specified axis.
# 【例】计算平均值(沿轴的元素的总和除以元素的数量)。
import numpy as np
x = np.array([[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]])
y = np.mean(x)
print(y)
y = np.mean(x,axis=0)
print(y)
y = np.mean(x, axis=1)
print(y)
23.0
[21. 22. 23. 24. 25.]
[13. 18. 23. 28. 33.]
计算加权平均值
- numpy.average(a[, axis=None, weights=None, returned=False])
Compute the weighted average along the specified axis.
mean 和average 都是计算均值的函数,在不指定权重的时候average 和mean 是一样的。指定权重后, average 可以计算加权平均值。
【例】计算加权平均值(将各数值乘以相应的权数,然后加总求和得到总体值,再除以总的单位数。)
import numpy as np
x = np.array([[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]])
y = np.average(x)
print(y)
print(np.mean(x))
y = np.average(x,axis=0)
print(y)
y = np.average(x,axis=1)
print(y)
y = np.arange(1,26).reshape([5,5])
print(y)
z = np.average(x,weights=y)
print(z)
z = np.average(x,axis=0,weights=y)
print(z)
z = np.average(x,axis=1,weights=y)
print(z)
23.0
23.0
[21. 22. 23. 24. 25.]
[13. 18. 23. 28. 33.]
[[ 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]]
27.0
[25.54545455 26.16666667 26.84615385 27.57142857 28.33333333]
[13.66666667 18.25 23.15384615 28.11111111 33.08695652]
[13.66666667 18.25 23.15384615 28.11111111 33.08695652]
计算方差
- numpy.var(a[, axis=None, dtype=None, out=None, ddof=0, keepdims=np._NoValue])
Computethe variance along the specified axis.
- ddof=0:是“Delta Degrees of Freedom”,表示自由度的个数。
要注意方差和样本方差的无偏估计,方差公式中分母上是n ;样本方差无偏估计公式中分母上是n‐1 ( n 为样本个数)。
# 【例】计算方差
import numpy as np
x = np.array([[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]])
y = np.var(x)
print(y)
y = np.mean((x - np.mean(x)) ** 2)
print(y)
y = np.var(x,ddof=1)
print(y)
y = np.var(x,axis=0)
print(y)
y = np.sum((x-np.mean(x))**2) / (x.size - 1)
print(y)
y = np.var(x,axis=0)
print(y)
y = np.var(x,axis=1)
print(y)
52.0
52.0
54.166666666666664
[50. 50. 50. 50. 50.]
54.166666666666664
[50. 50. 50. 50. 50.]
[2. 2. 2. 2. 2.]
计算标准差
numpy.std(a[, axis=None, dtype=None, out=None, ddof=0, keepdims=np._NoValue])
Compute the standard deviation along the specified axis.
标准差是一组数据平均值分散程度的一种度量,是方差的算术平方根。
import numpy as np
x = np.array([[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]])
y = np.std(x)
print(y)
y = np.sqrt(np.var(x))
print(y)
y = np.std(x,axis=0)
print(y)
y = np.std(x,axis=1)
print(y)
7.211102550927978
7.211102550927978
[7.07106781 7.07106781 7.07106781 7.07106781 7.07106781]
[1.41421356 1.41421356 1.41421356 1.41421356 1.41421356]
相关
计算协方差矩阵
- numpy.cov(m, y=None, rowvar=True, bias=False, ddof=None, fweights=None,aweights=None)
Estimate a covariance matrix, given data and weights.
# 【例】计算协方差矩阵
import numpy as np
x = [1,2,3,4,6]
y = [0,2,5,6,7]
print(np.cov(x))
print(np.cov(y))
print(np.cov(x,y))
print(np.var(x))
print(np.var(x,ddof=1))
print(np.var(y))
print(np.var(y,ddof=1))
z = np.mean((x-np.mean(x)) * (y - np.mean(y)))
print(z)
z = np.sum((x - np.mean(x)) * (y - np.mean(y)))/(len(x)-1)
print(z)
z = np.dot(x - np.mean(x),y - np.mean(y))/(len(x)-1)
print(z)
3.7
8.5
[[3.7 5.25]
[5.25 8.5 ]]
2.96
3.7
6.8
8.5
4.2
5.25
5.25
计算相关系数
- numpy.corrcoef(x, y=None, rowvar=True, bias=np._NoValue, ddof=np._NoValue)
Return Pearson product-moment correlation coefficients.
理解了np.cov() 函数之后,很容易理解np.correlate() ,二者参数几乎一模一样。
np.cov() 描述的是两个向量协同变化的程度,它的取值可能非常大,也可能非常小,这就导致没法直观地衡量二者协同变化的程度。相关系数实际上是正则化的协方差, n 个变量的相关系数形成一个n 维方阵。
# 【例】计算相关系数
import numpy as np
np.random.seed(20201127)
x, y = np.random.randint(0,20,size=[2,4])
print(x)
print(y)
z = np.corrcoef(x,y)
print(z)
a = np.dot(x-np.mean(x),y-np.mean(y))
b = np.sqrt(np.dot(x - np.mean(x),x-np.mean(x)))
c = np.sqrt(np.dot(y - np.mean(y),y-np.mean(y)))
print(a/(b*c))
[15 13 2 2]
[16 5 5 13]
[[1. 0.2465653]
[0.2465653 1. ]]
0.246565302221662
直方图
- numpy.digitize(x, bins, right=False)
Return the indices of the bins to which each value in input array belongs.
-
x:numpy数组
-
bins:一维单调数组,必须是升序或者降序
-
right:间隔是否包含最右
-
返回值:x在bins中的位置。
# 【例】
import numpy as np
x = np.array([0.2, 6.4, 3.0, 1.6])
bins = np.array([0.0, 1.0, 2.5, 4.0, 10.0])
inds = np.digitize(x,bins)
print(inds)
for n in range(x.size):
print(bins[inds[n]-1],'<=',x[n],'<',bins[inds[n]])
[1 4 3 2]
0.0 <= 0.2 < 1.0
4.0 <= 6.4 < 10.0
2.5 <= 3.0 < 4.0
1.0 <= 1.6 < 2.5
# 【例】
import numpy as np
x = np.array([1.2, 10.0, 12.4, 15.5, 20.])
bins = np.array([0, 5, 10, 15, 20])
inds = np.digitize(x, bins, right=True)
print(inds) # [1 2 3 4 4]
inds = np.digitize(x, bins, right=False)
print(inds)
[1 2 3 4 4]
[1 3 3 4 5]