#### pandas_dataframe1 ##########

import numpy as np
import pandas as pd

df = pd.DataFrame({
    'A': np.random.randint(1, 100, 4),
    'B': pd.date_range(start='20200401', periods=4, freq='D'),
    'C': pd.Series(range(4),
                   index=["zhang", "li", "zhou", "wang"],
                   dtype='float32'),
    'D': np.array([3] * 4, dtype='int32'),
    'E': pd.Categorical(["test", "train", "test", "train"]),
    'F': 'foo'
})

print(df)
# axis=0 纵向排序，axis=1 横向排序，ascending 默认为True，可以改为False
print(df.sort_index(axis=0,ascending=True))
print(df.sort_index(axis=0,ascending=False))
print(df.sort_index(axis=1,ascending=True))
print(df.sort_index(axis=1,ascending=False))

print(3 in df['C'])
print(3 in df['C'].values)

print(df[0:2])
print(df.loc[:,['A','C']])

print(df.at['zhang','A'])
print(df.loc[['zhang'],['A']])
print(df[df.E=='test'])
print(df[df['A'] > 50])
print(df[df['A'] > 50])
print(df.replace(3,5))
df1 = df.reindex(columns=list(df.columns) + ['G'])
print(df1)

df2 = df1['G'].fillna(5, inplace=True)
print(df2)


#### pandas_dataframe2 ##########
from idlelib.replace import replace

import numpy as np
import pandas as pd
from pandas.core.interchange.dataframe_protocol import DataFrame

data = pd.DataFrame(np.random.randn(500,4))
data.describe()

print(data.describe( ))

# # how = outer left right 不指定则默认为inner
# pd.merge(df1,df2,on='key',[how='outer' ])

df4 = pd.DataFrame({
    'A':np.random.randint(1,5,8),
    'B':np.random.randint(10,15,8),
    'C':np.random.randint(20,30,8),
    'D':np.random.randint(80,100,8),
    })

print(df4)

print(df4.groupby(['A','B'],as_index=False).mean())

print(df4.groupby(['A']).sum())
print(df4.groupby('A').sum())

# df4.to_excel
# pd.read_excel
# 数据清晰案例
df = pd.read_csv()
print(df)
print(df['num_ber'])
print(df['num_ber'].isnull())

# 定义缺失值，重要
missing_values = ['n/a','na','--']
# 统一转换为NaN
df = pd.read_csv('./test_data',na_values= missing_values)
df = pd.read_csv('./test_data',na_values= missing_values)
print(df)
print(df['aa'])
print(df['aa'].isnull())

new_df = df.dropna()
new_df2 = df.fillna(1234, inplace=True)
print(new_df.to_string())
df['st_num'].mean()
df['st_num'].fillna(1234, inplace=True)

# 日期格式化
df['Date'] = pd.to_datetime(df['Date'],format='mixed')

# 转换一个字典为Dataframe，里面加上字典变量
df = pd.DataFrame()

for x in df.dex:
    if df.loc[x,'age'] > 120:
        df.loc[x,'age'] = 120

# 去重复  里面不跟参数
df.duplicated()
df.drop_duplicates()



# 数据相关性分析  常用method有pearson和spearman
# pearson -1 1 区间内，靠近1 则正相关越大，靠近-1 则负相关越大
df.corr(method="pearson", min_periods=1)
# 不加任何参数直接调用
df.corr()


# 可视化：
# 直接调用
df.plot()
# plt.show()


#### pandas_demo1 ##########
from operator import index

import numpy as np
import pandas as pd
# x = pd.Series([1,3,5,np.nan])
# print(x)

# bb = range(5)
# for i in bb:
#     print(i)

# aa = pd.Series(range(5))
# print(aa.values)

# bb = pd.Series(range(5),index=list('abcde'))
# print(bb)
#
# cc = pd.date_range(start='20240101',end='20240131',freq='D')
# print(cc)

dates = pd.date_range(start='20240101',end='20240112',freq='D')
dd = pd.DataFrame(
    np.random.randn(12,4),
    index=dates,
    columns=list('ABCD')
)
# print(dd)
# print(dd.head())
# print(dd.head(3))
# print(dd.tail(1))


print(dd)
# print(dd['A'].index)
# print(dd['A'].keys)
# print(dd['A'])
for i in dd['A']:
    print(i)

# for i in dd['A'].values:
#     print(i)


#### pandas_demo1 ##########
import pandas as pd

missing_values = ['n/a', 'na', '--']

data = pd.read_csv("test_data.csv", na_values=missing_values)
print(data)
print("##################################")

# 删除两列 "Unnamed:0"和"Remark"
del data['Remark']
del data['Unnamed:0']
print(data)

print("##################################")
# 删除所有带NaN的行,并赋值给new_data
# print(data.dropna())
new_data = data.dropna()
print(new_data)
print("##################################")

# 日期格式化 用第一个没有报错，但是时间不对，用第二个数据是对的，但是有警告提醒。
# new_data.loc[:, 'Date'] = pd.to_datetime(new_data['Date'], format='mixed')
# new_data['Date'] = pd.to_datetime(new_data['Date'], format='mixed')
# new_data['Date'] = pd.to_datetime(new_data['Date'], unit='s')
new_data.loc[:, 'Date'] = pd.to_datetime(new_data['Date'], format='mixed')
# new_data.loc[:, 'Date'] = pd.to_datetime(new_data['Date'], unit='s')

# 去掉$
new_data.loc[:, 'Payment'] = new_data['Payment'].str.replace('$', '')
# 将，转为为 .
new_data.loc[:, 'Payment'] = new_data['Payment'].str.replace(',', '.')
# 转换为浮点型
new_data.loc[:, 'Payment'] = new_data['Payment'].astype(float)

# 删除Note里的乱七八糟的字符
new_data.loc[:, 'Note'] = new_data['Note'].str.replace('?', '')
new_data.loc[:, 'Note'] = new_data['Note'].str.replace('-', '')
new_data.loc[:, 'Note'] = new_data['Note'].str.replace('!', '')

# 分隔符倒序数值
new_data.loc[:, 'Name'] = new_data['Name'].str.split(',').apply(lambda x: ' '.join(x[::-1]))
# 转换为小写
new_data.loc[:, 'Name'] = new_data['Name'].str.lower()

print(new_data)


#### pandas_serires ##############
import numpy as np
import pandas as pd

# Series标准的“数据”实例化语法
demo_datas = pd.Series(range(5),index=list('abcde'))
print(demo_datas)

# Series标准的“日期”实例化语法 H小时  D天  M月
demo_dates = pd.date_range(start='20240101',end='20240303',freq='D')
print(demo_dates)

# 怎么用用Series生成DataFrame？



x = pd.Series(['1,3,5,np.nan'])

# print(x.info)
# print(x.values)

#### test1 ##############
coding: utf-8

num = -5
if num > 0:
    print("num为正数")
else:
    print("num为负数")
print(abs(num))


nums = [1,2,3,4,5]
# print(sorted(nums,reverse=True))
print(nums.__reversed__())

help()
#
# print(max(nums))
a = 10
b = 3
print(round(a/b,2))
print(pow(a,b))


#
# for i in nums:
#     print(i)
#
# total = 0
#
# while True:
#     num = int(input("请输入一个数："))
#     total = total + num
#     if num == 0:
#         break
#
# print("综合为：",total)


import random as rd

aa = rd.randint(10)

print(aa)


nums = [1, 2, 3, 4, 5]

def test():
    sum = 0
    for i in nums:
        sum = sum + i
    return sum / len(nums)

print(test())


#### test2 ##############
# coding: utf-8

import numpy as np

matrix = np.random.rand(55)
print(matrix)
print(np.mean(matrix))
print(np.sum(matrix))
print(sum(sum(matrix)))




def test(param1):
    print(param1)
    return True

# test(123)
print(test("aa"))

def test2(param1,param2):
    return True
print(test2(1,2))

for i in range(1,100,2):
    print(i)



#### test3 ##############
class Rectangole:
    def __init__(self, width, height):
        self.width = width
        self.height = height

    def area(self):
        return self.width * self.height


my_rect = Rectangole(5, 10)
print(my_rect.area())


class Animal:
    def __init__(self, name):
        self.name = name

    def run(self):
        print(self.name + " is running")


class Dog(Animal):
    def bark(self):
        print(self.name + " is barking")


d = Dog("Jack")
d.run()
d.bark()

########### test_data.csv ##############

Unnamed:0,Date,Name,Payment,Note,Remark
0,04/04/2021,"doe,john","$100,50",Unhappy!,
1,"april 17th,2021","Doe,Jane",$78.50 ,?Satidfied,
2,2021-8-21,"smith,Adam",$65 ,Neutral,
3,05/02/2021,"Tuck,matt",$120 ,Unhappy-,
4,05/12/2021,"James,Ben",,Neutral0,
5,2021-10-11,,$100 ,Neutral,
6,,,,,