# -*- coding: utf-8 -*-
"""
Created on Sun Sep 3 16:30:59 2023
"""
import numpy as np
import pandas as pd
# =======================
# list
a = [2,3,1,'1',5] #list()
a.append(1) #加
print(a)
a.remove(1) #值
print(a)
b = a.pop(0) #位置 a.pop() 结尾元素弹出并赋值
print(a, b)
b = sorted(a)
print(a, b)
order = sorted(range(len(a)), key=lambda i: a[i]) #索引排序
print(order, np.array(a)[order])
a.sort() #直接对a排序
print(a)
print(a[1], a[-1], a[1:], a[:-1])
print(a[:1] + a[-2:])
print([1]*3)
b = a.copy() #a[:]
b[1] = 1
print(a, b)
[i**2 for i in range(5)]
[(i,ai) for i,ai in enumerate(a)]
# =======================
# dict - dict内元素无序，不一定先入就排在前面
a = {'x':1, 'y':2} #dict()
a['z'] = 3
print(a)
a['x'] = 0
print(a)
del a['z']
print(a, a.keys(), a.values(), a.items())
{ key:val**2 for key,val in a.items() }
# =======================
# tuple 元组 有序，元素不可变，可以用元组定义dict的key, 但是list不行
a = tuple([1,2,2,3,'3'])
print(a)
print(a[:1] + a[-2:])
# =======================
# 集合 set，元素无重复，无序
a = set([1,2,2,3,'3'])
print(a)
a.add('4')
print(a)
a.pop() #第一个元素弹出
a.remove(1)
a.difference(set([1,2])) #差集
a.intersection(set([3,'3',6])) #交集
a.union(set([1,2])) #交集
a.issubset(set([2,3]))
a.issuperset(set([2,3]))
#string
a = '23'
a+a
a*3
list(a)
[i*2 for i in a]
','.join(list(a))
print('ab%sd' % 'c')
print('ab%s%sd' % ('c', 'f'))
print('ab{}{}d'.format('c', 'f'))
print(f'ab{a}{a}')
b = '''
as
'''
print(b)
# =======================
# numpy
# =======================
a = np.array([1,2,3])
np.ones_like(a)
np.zeros_like(a)
np.empty_like(a)
a = np.arange(5)
np.linspace(-1, 1, 100)
a = np.random.randn(5)
a.round(2)
a.astype(int)
a.cumsum()
np.prod(a)
np.random.rand(3,4)
np.ones(3)
np.zeros(3,3)
np.diag([[2,3],[1,4]])
np.diag([2,3])
a = np.random.randn(5)
np.sum(a)
a.sum()
a.mean()
a.std(ddof=1)
a.median()
np.percentile(a, [25,50,75])
a = np.random.randn(3,3)
a+a
a-a
a*a
a/a
a**2
a @ a #np.sum(a * a)
np.log(a)
np.exp(a)
np.pi
np.abs(a)
np.max(a)
np.min(a)
np.argmax(a)
np.argmin(a)
np.where(a>0) #原始数据几维，返回的tuple就为几维，可以【0】取出元素值为True的第0维索引
np.where(a>0, 1, -1)
np.all(a > 0)
np.any(a > 0)
np.r_[a,a]
np.c_[a,a]
np.concatenate((a, a), axis=0)
for i in a:
print(i)
# =======================
# pandas
# =======================
path = r'd:\test.csv' #'d:/test.csv'
pd.read_csv(path, encoding='utf-8')
pd.to_csv(path, index=False)
pd.read_clipboard() #header=0, sep='\t'
a = pd.Series([1.234,2,3])
a.index
a.values
a.round(2)
a.map(lambda x: x**2)
dict(a)
a = pd.DataFrame({'x': [1,2],
'y': [3,4]})
a = pd.DataFrame([[1,2],[3,4]], columns=['x','y'])
a.index
a.columns #可直接赋值改变列名
a.values
a.head(5)
a.sum(axis=0)
a.mean()
a.std()
a.apply(lambda ser: ser['x'], axis=1)
a['x'] #series
a[['x']] #data frame
a.x
a.loc[:, 'x']
a.iloc[:, 0]
a['z'] = [3,5]
del a['z']
for i,ser in a.iterrows():
print(i, ser)
