【Pandas】入门案例

 1 # To add a new cell, type ‘# %%‘
 2 # To add a new markdown cell, type ‘# %% [markdown]‘
 3 # %%
 4 import numpy as np
 5 import pandas as pd
 6 # 使用 Series 生产序列，Pandas默认生成整数索引
 7 s = pd.Series([1,3,4, np.nan, 6,8]) 
 8 s
 9 
10 
11 # %%
12 # 使用含日期时间索引和标签的Numpy数组生成DateFrame
13 dates = pd.date_range(‘20200703‘, periods = 6)
14 dates
15 df = pd.DataFrame(np.random.randn(6,5),index = dates, columns = list(‘ABCDE‘))
16 df
17 
18 
19 # %%
20 import pandas as pd
21 import numpy as np
22 # 使用Series字典对象生成DataFrame
23 df2 = pd.DataFrame({"A": 1.,
24                     "N": pd.Timestamp(‘20200703‘),
25                     "B": pd.Series(1, index = list(range(4)),dtype=‘float32‘),
26                     "D": np.array([3]*4,dtype = ‘int32‘),
27                     "E": pd.Categorical(["test","train","test","train"]),
28                     "F": ‘foo‘})
29 df2 
30 
31 
32 # %%
33 #查询数据类型
34 df2.dtypes
35 
36 
37 # %%
38 # 获取单列数据
39 df2.A
40 
41 
42 # %%
43 df2.append
44 
45 
46 # %%
47 df.head() #Series 或着 DataFrame 前5位
48 
49 
50 # %%
51 df.tail(3) #Series或DataFrame 的尾部数据 可选数量
52 
53 
54 # %%
55 ts = pd.Series(np.random.randn(1000),
56                index = pd.date_range(‘1/1/2000‘,periods=1000))
57 ts = ts.cumsum()
58 ts.plot()
59 
60 
61 # %%
62 df.index
63 
64 
65 # %%
66 df.columns
67 
68 
69 # %%
70 df.to_numpy()
71 
72 
73 # %%
74 df2.to_numpy()
75 
76 
77 # %%
78 df.describe()
79 
80 
81 # %%
82 df.T
83 
84 
85 # %%
86 df.sort_index(axis=1,ascending=False)
87 
88 
89 # %%
90 df.sort_values(by=‘B‘)
91 
92 
93 # %%