- Counter函数可以对列表中数据进行统计每一个有多少种
- most_common(10)可以提取前十位
from collections import Counter
a = ['q','q','w','w','w']
count = Counter(a)
count.most_common(1)
[('w', 3)]
count
Counter({'q': 2, 'w': 3})
-
pandas中的series对象有一个value_counts方法可以计数
-
.fillna()函数可以替换确实值NA
import numpy as np
from numpy.random import randn
data = {i : randn() for i in range(7)}
data
{0: -0.2657989059225722,
1: -1.2517286143172295,
2: -0.6360811023039581,
3: 1.2009891917346602,
4: 1.7528414640242418,
5: -0.24155970563487628,
6: -0.7637924413712933}
- 最近的两个结果保存在_和__中
9*3
27
_
27
%pwd
# 获得当前工作目录
'D:\Code\Python\code'
- 魔法命令
- %time 一条语句的执行时间
- %timeit 执行多次的平均时间
numpy基础
import numpy as np
data1 = [6, 7.5, 8],[2, 0 ,1]
arr1 = np.array(data1)
arr1
array([[6. , 7.5, 8. ],
[2. , 0. , 1. ]])
arr1.ndim
2
arr1.shape
(2, 3)
arr1.dtype
dtype('float64')
np.zeros((2, 3, 4))
array([[[0., 0., 0., 0.],
[0., 0., 0., 0.],
[0., 0., 0., 0.]],
[[0., 0., 0., 0.],
[0., 0., 0., 0.],
[0., 0., 0., 0.]]])
np.ones((3))
array([1., 1., 1.])
np.arange(9)
array([0, 1, 2, 3, 4, 5, 6, 7, 8])
np.eye(3,)
array([[1., 0., 0.],
[0., 1., 0.],
[0., 0., 1.]])
- 数组切片后的修改会反映到原始数组上
arr = np.arange(9)
arr2 = arr[5:8]
arr2[:] = 4
arr
array([0, 1, 2, 3, 4, 4, 4, 4, 8])
arr = np.arange(9)
arr2 = arr[5:8].copy()
arr2[:] = 4
arr
array([0, 1, 2, 3, 4, 5, 6, 7, 8])
- 三维数组先是层、行、列
data = np.random.randn(4,3)
data
array([[ 0.7144327 , 0.87144603, 1.10651404],
[-0.19509352, -0.01102958, 1.8051039 ],
[ 0.03106339, 0.83767495, 0.20094192],
[ 0.96032146, -0.40303045, 1.4522938 ]])
data[[1,2,0],[1,2,0]]
# 取出来的数据为（11）（22）（00）
array([-0.01102958, 0.20094192, 0.7144327 ])
- 不连续提取数据
data[[0,2]][:,[0,2]]
# 跳着取方法一
array([[0.7144327 , 1.10651404],
[0.03106339, 0.20094192]])
data[np.ix_([0,2],[0,2])]
# 跳着取方法二
array([[0.7144327 , 1.10651404],
[0.03106339, 0.20094192]])
data.T
array([[ 0.7144327 , -0.19509352, 0.03106339, 0.96032146],
[ 0.87144603, -0.01102958, 0.83767495, -0.40303045],
[ 1.10651404, 1.8051039 , 0.20094192, 1.4522938 ]])
np.sqrt(data)
F:Anacondalibsite-packagesipykernel_launcher.py:1: RuntimeWarning: invalid value encountered in sqrt
"""Entry point for launching an IPython kernel.
array([[0.84524121, 0.93351274, 1.05190971],
[ nan, nan, 1.34354155],
[0.17624808, 0.91524584, 0.44826546],
[0.97995993, nan, 1.20511153]])
np.exp(data)
array([[2.04302734, 2.39036489, 3.02379915],
[0.82275771, 0.98903102, 6.0806032 ],
[1.03155089, 2.31098757, 1.22255377],
[2.61253617, 0.66829175, 4.27290447]])
np.rint(data)
# 四舍五入
array([[ 1., 1., 1.],
[-0., -0., 2.],
[ 0., 1., 0.],
[ 1., -0., 1.]])
np.modf(data)
# 将数据分为小数和整数部分
(array([[ 0.7144327 , 0.87144603, 0.10651404],
[-0.19509352, -0.01102958, 0.8051039 ],
[ 0.03106339, 0.83767495, 0.20094192],
[ 0.96032146, -0.40303045, 0.4522938 ]]), array([[ 0., 0., 1.],
[-0., -0., 1.],
[ 0., 0., 0.],
[ 0., -0., 1.]]))
np.isnan(data)
array([[False, False, False],
[False, False, False],
[False, False, False],
[False, False, False]])
np.where(data > 0,9,data)
array([[ 9. , 9. , 9. ],
[-0.19509352, -0.01102958, 9. ],
[ 9. , 9. , 9. ],
[ 9. , -0.40303045, 9. ]])
- axis中0表示竖向求和，1表示横向求和
np.mean(data,axis=1)
array([0.89746426, 0.5329936 , 0.35656009, 0.6698616 ])
np.in1d(data, [1,2,3])
# 查看data中每个元素是否在1，2，3内
array([False, False, False, False, False, False, False, False, False,
False, False, False])
from numpy.linalg import inv, qr
from numpy.random import randn
x = randn(5,5)
mat = x.dot(inv(x))
# 求逆
mat = np.rint(mat)
mat
array([[ 1., 0., -0., 0., 0.],
[ 0., 1., -0., 0., 0.],
[ 0., -0., 1., 0., 0.],
[-0., 0., -0., 1., -0.],
[-0., 0., 0., -0., 1.]])
np.diag(mat)
# 返回对角线元素
array([1., 1., 1., 1., 1.])
np.random.permutation(mat)
# 返回序列的随机排列
array([[ 0., -0., 1., 0., 0.],
[-0., 0., -0., 1., -0.],
[ 1., 0., -0., 0., 0.],
[-0., 0., 0., -0., 1.],
[ 0., 1., -0., 0., 0.]])
np.random.randint(0,2,12)
array([1, 1, 1, 0, 0, 1, 1, 0, 1, 1, 1, 1])