DataFrame的结构

import pandas as pd
import numpy as np

# 创建一个符合正态分布的10个股票的5天的涨跌幅数据
stock_change=np.random.normal(0,1,(10,5))

stock_change

array([[ 1.76548834, -2.1033278 , -0.44765295,  1.6133684 ,  0.56440317],
       [-0.28406163,  0.33190268,  0.15697138, -0.15352828,  0.39168077],
       [ 0.97804543,  0.70247765,  0.48852276, -0.84483313, -0.4901491 ],
       [ 0.32663842, -1.18491493, -1.85617695, -0.08334347, -0.88467526],
       [-0.0606477 , -0.3282171 ,  0.13469079, -0.33644424,  0.49229211],
       [-0.08971546,  0.29502656,  0.58225254, -1.15526343,  0.1121633 ],
       [ 0.08384949,  0.63896248,  0.03189999, -0.17538923, -0.06095104],
       [ 0.0892971 , -1.54426915,  0.22247612, -0.31827644,  1.4569718 ],
       [-1.31604909, -1.09219638, -0.64118638, -0.55174267,  2.54943429],
       [-0.75827099,  0.19581938, -0.11856612,  0.44903057, -0.07891904]])

# 创建dataframe
stock_num=["股票{}".format(i)for i in range(10)]  #构造行标
date=pd.date_range(start="20200629", periods=5,freq="B")

stock_change=pd.DataFrame(stock_change, index=stock_num,columns=date)
stock_change

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

stock_change.shape

(10, 5)

stock_change.index

Index(['股票0', '股票1', '股票2', '股票3', '股票4', '股票5', '股票6', '股票7', '股票8', '股票9'], dtype='object')

stock_change.columns

DatetimeIndex(['2020-06-29', '2020-06-30', '2020-07-01', '2020-07-02',
               '2020-07-03'],
              dtype='datetime64[ns]', freq='B')

stock_change.values

array([[ 1.76548834, -2.1033278 , -0.44765295,  1.6133684 ,  0.56440317],
       [-0.28406163,  0.33190268,  0.15697138, -0.15352828,  0.39168077],
       [ 0.97804543,  0.70247765,  0.48852276, -0.84483313, -0.4901491 ],
       [ 0.32663842, -1.18491493, -1.85617695, -0.08334347, -0.88467526],
       [-0.0606477 , -0.3282171 ,  0.13469079, -0.33644424,  0.49229211],
       [-0.08971546,  0.29502656,  0.58225254, -1.15526343,  0.1121633 ],
       [ 0.08384949,  0.63896248,  0.03189999, -0.17538923, -0.06095104],
       [ 0.0892971 , -1.54426915,  0.22247612, -0.31827644,  1.4569718 ],
       [-1.31604909, -1.09219638, -0.64118638, -0.55174267,  2.54943429],
       [-0.75827099,  0.19581938, -0.11856612,  0.44903057, -0.07891904]])

stock_change.T

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

stock_change.head(7)

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

# 索引值修改
new_index=["股票00{}".format(i)for i in range(10)] 
stock_change.index=new_index

stock_change

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

stock_change.reset_index(drop=True)

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

stock_change

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

stock_change.set_index("2020-6-29 00:00:00")

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

stock_change

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

type(stock_change.loc["股票004"])

pandas.core.series.Series

stock_change.loc["股票004"]

2020-06-29   -0.060648
2020-06-30   -0.328217
2020-07-01    0.134691
2020-07-02   -0.336444
2020-07-03    0.492292
Freq: B, Name: 股票004, dtype: float64

pd.Series(np.arange(2,20,2))

0     2
1     4
2     6
3     8
4    10
5    12
6    14
7    16
8    18
dtype: int64

sr=pd.Series({'red':100,"blue":200,"queen":500, 'yellow':1000})

sr

blue       200
queen      500
red        100
yellow    1000
dtype: int64

sr.index

Index(['blue', 'queen', 'red', 'yellow'], dtype='object')

sr.values

array([ 200,  500,  100, 1000])

pandas 操作suoyin

data=pd.read_csv("./stock_day/stock_day.csv")

data = data.drop(["ma5","ma10","ma20","v_ma5","v_ma10","v_ma20"], axis=1)  #让数据简单一些

data.head()

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

data["open"]["2018-02-23"]   #pandas必须先列后行

22.88

data.loc["2018-02-23"]["open"]  #使用loc可以先行后列

22.88

data.loc["2018-02-26","open"]

22.8

data.iloc[1,0]

22.8

data.loc[data.index[0:4],['open','close','high','low']]

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

data.iloc[0:4,data.columns.get_indexer(['open','close','high','low'])]

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

data.price_change.head()

2018-02-27    0.63
2018-02-26    0.69
2018-02-23    0.54
2018-02-22    0.36
2018-02-14    0.44
Name: price_change, dtype: float64

# data["open"]=8

data.head()

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

data=data.sort_values(by="p_change", ascending=False).head()
data.head()

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

data=data.sort_values(by=['open','high'])

data.head()

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

data.sort_index()

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

data["p_change"].sort_values(ascending=True)

2015-08-04    10.02
2016-12-22    10.02
2016-07-07    10.02
2015-05-21    10.02
2015-08-28    10.03
Name: p_change, dtype: float64

data["p_change"].sort_index(ascending=True)

2015-05-21    10.02
2015-08-04    10.02
2015-08-28    10.03
2016-07-07    10.02
2016-12-22    10.02
Name: p_change, dtype: float64

data

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

data+100

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

data/100

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

data.add(100)

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

data["p_change"]>2

2015-08-28    True
2015-08-04    True
2016-12-22    True
2016-07-07    True
2015-05-21    True
Name: p_change, dtype: bool

data

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

data[data["open"]>17]

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

data.query("p_change>10&open>18").head()

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

data["price_change"].isin([1.86,2.57])

2015-08-28    False
2015-08-04    False
2016-12-22     True
2016-07-07    False
2015-05-21     True
Name: price_change, dtype: bool

data.describe()

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

data.idxmax()

open            2015-05-21
high            2015-05-21
close           2015-05-21
low             2015-05-21
volume          2016-12-22
price_change    2015-05-21
p_change        2015-08-28
turnover        2015-05-21
dtype: object

data=pd.read_csv("./stock_day/stock_day.csv")
data = data.drop(["ma5","ma10","ma20","v_ma5","v_ma10","v_ma20"], axis=1)  #让数据简单一些

data.head()

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

data["p_change"].sort_index().cumsum().head()

2015-03-02     2.62
2015-03-03     4.06
2015-03-04     5.63
2015-03-05     7.65
2015-03-06    16.16
Name: p_change, dtype: float64

data["p_change"].sort_index().cumsum().plot()

<matplotlib.axes._subplots.AxesSubplot at 0x7f6c86b3c9e8>