CC BY 4.0 （除特别声明或转载文章外）
如果这篇博客帮助到你，可以请我喝一杯咖啡~
Anaconda简介
本教程来源于学堂在线
MacOS下部分命令
-
创建名为ml的python3.6环境
conda create -n ml python=3.6 -
创建名为ml的环境
conda env remove -n ml -
进入名为ml的环境
source activate ml -
退出环境
conda deactivate -
安装工具包
conda install numpy pandas scikit-learn -
查看所有环境名字
conda env list -
查看当前环境下所有已安装的工具包
conda list
Pandas & Numpy
为帮助python基础较弱的同学完成案例作业，下面为大家演示讲解numpy和pandas里的一些常用函数，了解基本操作方便之后使用。
载入工具包
import pandas as pd
import numpy as np
Pandas
初始化一个pandas的DataFrame
比较常用的一种初始化方式是从python的字典初始化，列名是字典的key，每列的元素是字典的value，要求每列长度相同
df = pd.DataFrame({'age': [1,2,3], 'name': ['a', 'b', 'c']})
print(df)
print(type(df))
age name
0 1 a
1 2 b
2 3 c
<class 'pandas.core.frame.DataFrame'>
pandas也可以读入一个csv文件成DataFrame
df = pd.read_csv('./data/high_diamond_ranked_10min.csv', sep=',')
print(type(df))
<class 'pandas.core.frame.DataFrame'>
查看DataFrame的信息
df.info()
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 9879 entries, 0 to 9878
Data columns (total 40 columns):
gameId 9879 non-null int64
blueWins 9879 non-null int64
blueWardsPlaced 9879 non-null int64
blueWardsDestroyed 9879 non-null int64
blueFirstBlood 9879 non-null int64
blueKills 9879 non-null int64
blueDeaths 9879 non-null int64
blueAssists 9879 non-null int64
blueEliteMonsters 9879 non-null int64
blueDragons 9879 non-null int64
blueHeralds 9879 non-null int64
blueTowersDestroyed 9879 non-null int64
blueTotalGold 9879 non-null int64
blueAvgLevel 9879 non-null float64
blueTotalExperience 9879 non-null int64
blueTotalMinionsKilled 9879 non-null int64
blueTotalJungleMinionsKilled 9879 non-null int64
blueGoldDiff 9879 non-null int64
blueExperienceDiff 9879 non-null int64
blueCSPerMin 9879 non-null float64
blueGoldPerMin 9879 non-null float64
redWardsPlaced 9879 non-null int64
redWardsDestroyed 9879 non-null int64
redFirstBlood 9879 non-null int64
redKills 9879 non-null int64
redDeaths 9879 non-null int64
redAssists 9879 non-null int64
redEliteMonsters 9879 non-null int64
redDragons 9879 non-null int64
redHeralds 9879 non-null int64
redTowersDestroyed 9879 non-null int64
redTotalGold 9879 non-null int64
redAvgLevel 9879 non-null float64
redTotalExperience 9879 non-null int64
redTotalMinionsKilled 9879 non-null int64
redTotalJungleMinionsKilled 9879 non-null int64
redGoldDiff 9879 non-null int64
redExperienceDiff 9879 non-null int64
redCSPerMin 9879 non-null float64
redGoldPerMin 9879 non-null float64
dtypes: float64(6), int64(34)
memory usage: 3.0 MB
查看头/尾10条数据
# df.head(10)
# df.tail(10)
print(df)
print(df.drop(columns=['gameId']))
print(df)
gameId blueWins blueWardsPlaced blueWardsDestroyed \
0 4519157822 0 28 2
1 4523371949 0 12 1
2 4521474530 0 15 0
3 4524384067 0 43 1
4 4436033771 0 75 4
... ... ... ... ...
9874 4527873286 1 17 2
9875 4527797466 1 54 0
9876 4527713716 0 23 1
9877 4527628313 0 14 4
9878 4523772935 1 18 0
blueFirstBlood blueKills blueDeaths blueAssists blueEliteMonsters \
0 1 9 6 11 0
1 0 5 5 5 0
2 0 7 11 4 1
3 0 4 5 5 1
4 0 6 6 6 0
... ... ... ... ... ...
9874 1 7 4 5 1
9875 0 6 4 8 1
9876 0 6 7 5 0
9877 1 2 3 3 1
9878 1 6 6 5 0
blueDragons ... redTowersDestroyed redTotalGold redAvgLevel \
0 0 ... 0 16567 6.8
1 0 ... 1 17620 6.8
2 1 ... 0 17285 6.8
3 0 ... 0 16478 7.0
4 0 ... 0 17404 7.0
... ... ... ... ... ...
9874 1 ... 0 15246 6.8
9875 1 ... 0 15456 7.0
9876 0 ... 0 18319 7.4
9877 1 ... 0 15298 7.2
9878 0 ... 0 15339 6.8
redTotalExperience redTotalMinionsKilled redTotalJungleMinionsKilled \
0 17047 197 55
1 17438 240 52
2 17254 203 28
3 17961 235 47
4 18313 225 67
... ... ... ...
9874 16498 229 34
9875 18367 206 56
9876 19909 261 60
9877 18314 247 40
9878 17379 201 46
redGoldDiff redExperienceDiff redCSPerMin redGoldPerMin
0 -643 8 19.7 1656.7
1 2908 1173 24.0 1762.0
2 1172 1033 20.3 1728.5
3 1321 7 23.5 1647.8
4 1004 -230 22.5 1740.4
... ... ... ... ...
9874 -2519 -2469 22.9 1524.6
9875 -782 -888 20.6 1545.6
9876 2416 1877 26.1 1831.9
9877 839 1085 24.7 1529.8
9878 -927 58 20.1 1533.9
[9879 rows x 40 columns]
blueWins blueWardsPlaced blueWardsDestroyed blueFirstBlood \
0 0 28 2 1
1 0 12 1 0
2 0 15 0 0
3 0 43 1 0
4 0 75 4 0
... ... ... ... ...
9874 1 17 2 1
9875 1 54 0 0
9876 0 23 1 0
9877 0 14 4 1
9878 1 18 0 1
blueKills blueDeaths blueAssists blueEliteMonsters blueDragons \
0 9 6 11 0 0
1 5 5 5 0 0
2 7 11 4 1 1
3 4 5 5 1 0
4 6 6 6 0 0
... ... ... ... ... ...
9874 7 4 5 1 1
9875 6 4 8 1 1
9876 6 7 5 0 0
9877 2 3 3 1 1
9878 6 6 5 0 0
blueHeralds ... redTowersDestroyed redTotalGold redAvgLevel \
0 0 ... 0 16567 6.8
1 0 ... 1 17620 6.8
2 0 ... 0 17285 6.8
3 1 ... 0 16478 7.0
4 0 ... 0 17404 7.0
... ... ... ... ... ...
9874 0 ... 0 15246 6.8
9875 0 ... 0 15456 7.0
9876 0 ... 0 18319 7.4
9877 0 ... 0 15298 7.2
9878 0 ... 0 15339 6.8
redTotalExperience redTotalMinionsKilled redTotalJungleMinionsKilled \
0 17047 197 55
1 17438 240 52
2 17254 203 28
3 17961 235 47
4 18313 225 67
... ... ... ...
9874 16498 229 34
9875 18367 206 56
9876 19909 261 60
9877 18314 247 40
9878 17379 201 46
redGoldDiff redExperienceDiff redCSPerMin redGoldPerMin
0 -643 8 19.7 1656.7
1 2908 1173 24.0 1762.0
2 1172 1033 20.3 1728.5
3 1321 7 23.5 1647.8
4 1004 -230 22.5 1740.4
... ... ... ... ...
9874 -2519 -2469 22.9 1524.6
9875 -782 -888 20.6 1545.6
9876 2416 1877 26.1 1831.9
9877 839 1085 24.7 1529.8
9878 -927 58 20.1 1533.9
[9879 rows x 39 columns]
gameId blueWins blueWardsPlaced blueWardsDestroyed \
0 4519157822 0 28 2
1 4523371949 0 12 1
2 4521474530 0 15 0
3 4524384067 0 43 1
4 4436033771 0 75 4
... ... ... ... ...
9874 4527873286 1 17 2
9875 4527797466 1 54 0
9876 4527713716 0 23 1
9877 4527628313 0 14 4
9878 4523772935 1 18 0
blueFirstBlood blueKills blueDeaths blueAssists blueEliteMonsters \
0 1 9 6 11 0
1 0 5 5 5 0
2 0 7 11 4 1
3 0 4 5 5 1
4 0 6 6 6 0
... ... ... ... ... ...
9874 1 7 4 5 1
9875 0 6 4 8 1
9876 0 6 7 5 0
9877 1 2 3 3 1
9878 1 6 6 5 0
blueDragons ... redTowersDestroyed redTotalGold redAvgLevel \
0 0 ... 0 16567 6.8
1 0 ... 1 17620 6.8
2 1 ... 0 17285 6.8
3 0 ... 0 16478 7.0
4 0 ... 0 17404 7.0
... ... ... ... ... ...
9874 1 ... 0 15246 6.8
9875 1 ... 0 15456 7.0
9876 0 ... 0 18319 7.4
9877 1 ... 0 15298 7.2
9878 0 ... 0 15339 6.8
redTotalExperience redTotalMinionsKilled redTotalJungleMinionsKilled \
0 17047 197 55
1 17438 240 52
2 17254 203 28
3 17961 235 47
4 18313 225 67
... ... ... ...
9874 16498 229 34
9875 18367 206 56
9876 19909 261 60
9877 18314 247 40
9878 17379 201 46
redGoldDiff redExperienceDiff redCSPerMin redGoldPerMin
0 -643 8 19.7 1656.7
1 2908 1173 24.0 1762.0
2 1172 1033 20.3 1728.5
3 1321 7 23.5 1647.8
4 1004 -230 22.5 1740.4
... ... ... ... ...
9874 -2519 -2469 22.9 1524.6
9875 -782 -888 20.6 1545.6
9876 2416 1877 26.1 1831.9
9877 839 1085 24.7 1529.8
9878 -927 58 20.1 1533.9
[9879 rows x 40 columns]
DataFrame每行的index
这里的index就是普通的数字0-9879，有时index也可以是其他一些特殊对象，如日期时间等。
print(df.index)
RangeIndex(start=0, stop=9879, step=1)
列名
csv文件一般列名读入是字符串str，有些情况下没有列名或简单以数字作为列名。
print(df.columns)
cols = df.columns[:2]
print(type(cols))
cols = list(cols) + ['blueAvgLevel', 'redAvgLevel']
print(cols)
Index(['gameId', 'blueWins', 'blueWardsPlaced', 'blueWardsDestroyed',
'blueFirstBlood', 'blueKills', 'blueDeaths', 'blueAssists',
'blueEliteMonsters', 'blueDragons', 'blueHeralds',
'blueTowersDestroyed', 'blueTotalGold', 'blueAvgLevel',
'blueTotalExperience', 'blueTotalMinionsKilled',
'blueTotalJungleMinionsKilled', 'blueGoldDiff', 'blueExperienceDiff',
'blueCSPerMin', 'blueGoldPerMin', 'redWardsPlaced', 'redWardsDestroyed',
'redFirstBlood', 'redKills', 'redDeaths', 'redAssists',
'redEliteMonsters', 'redDragons', 'redHeralds', 'redTowersDestroyed',
'redTotalGold', 'redAvgLevel', 'redTotalExperience',
'redTotalMinionsKilled', 'redTotalJungleMinionsKilled', 'redGoldDiff',
'redExperienceDiff', 'redCSPerMin', 'redGoldPerMin'],
dtype='object')
<class 'pandas.core.indexes.base.Index'>
['gameId', 'blueWins', 'blueAvgLevel', 'redAvgLevel']
访问列
可以传入列名字符串（返回Series）或列名的list（返回DataFame）
df['redAvgLevel']
# df[cols]
type(df[['redAvgLevel']])
pandas.core.frame.DataFrame
访问行
df[0:3]
| gameId | blueWins | blueWardsPlaced | blueWardsDestroyed | blueFirstBlood | blueKills | blueDeaths | blueAssists | blueEliteMonsters | blueDragons | ... | redTowersDestroyed | redTotalGold | redAvgLevel | redTotalExperience | redTotalMinionsKilled | redTotalJungleMinionsKilled | redGoldDiff | redExperienceDiff | redCSPerMin | redGoldPerMin | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 4519157822 | 0 | 28 | 2 | 1 | 9 | 6 | 11 | 0 | 0 | ... | 0 | 16567 | 6.8 | 17047 | 197 | 55 | -643 | 8 | 19.7 | 1656.7 |
| 1 | 4523371949 | 0 | 12 | 1 | 0 | 5 | 5 | 5 | 0 | 0 | ... | 1 | 17620 | 6.8 | 17438 | 240 | 52 | 2908 | 1173 | 24.0 | 1762.0 |
| 2 | 4521474530 | 0 | 15 | 0 | 0 | 7 | 11 | 4 | 1 | 1 | ... | 0 | 17285 | 6.8 | 17254 | 203 | 28 | 1172 | 1033 | 20.3 | 1728.5 |
3 rows × 40 columns
访问某个区域
例如前3行，cols对应的4列
df.loc[range(0,3), cols]
| gameId | blueWins | blueAvgLevel | redAvgLevel | |
|---|---|---|---|---|
| 0 | 4519157822 | 0 | 6.6 | 6.8 |
| 1 | 4523371949 | 0 | 6.6 | 6.8 |
| 2 | 4521474530 | 0 | 6.4 | 6.8 |
不建议在loc中使用形如0:3这样的index，因为行为略为反常。
df.loc[0:3]
| gameId | blueWins | blueWardsPlaced | blueWardsDestroyed | blueFirstBlood | blueKills | blueDeaths | blueAssists | blueEliteMonsters | blueDragons | ... | redTowersDestroyed | redTotalGold | redAvgLevel | redTotalExperience | redTotalMinionsKilled | redTotalJungleMinionsKilled | redGoldDiff | redExperienceDiff | redCSPerMin | redGoldPerMin | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 4519157822 | 0 | 28 | 2 | 1 | 9 | 6 | 11 | 0 | 0 | ... | 0 | 16567 | 6.8 | 17047 | 197 | 55 | -643 | 8 | 19.7 | 1656.7 |
| 1 | 4523371949 | 0 | 12 | 1 | 0 | 5 | 5 | 5 | 0 | 0 | ... | 1 | 17620 | 6.8 | 17438 | 240 | 52 | 2908 | 1173 | 24.0 | 1762.0 |
| 2 | 4521474530 | 0 | 15 | 0 | 0 | 7 | 11 | 4 | 1 | 1 | ... | 0 | 17285 | 6.8 | 17254 | 203 | 28 | 1172 | 1033 | 20.3 | 1728.5 |
| 3 | 4524384067 | 0 | 43 | 1 | 0 | 4 | 5 | 5 | 1 | 0 | ... | 0 | 16478 | 7.0 | 17961 | 235 | 47 | 1321 | 7 | 23.5 | 1647.8 |
4 rows × 40 columns
访问某个元素
loc是传入index和列名， 但是iloc传入的是编号，无论index和column是否为数字，都传入0-xxx的数字下标
df.loc[0, 'gameId']
df.iloc[0, 0]
df.at[0, 'gameId']
4519157822
拷贝
=是引用一个DataFrame对象，修改df_copy则df也会发生变化。 如果不想原df被修改，可以使用copy深拷贝一个DataFrame对象。
df_copy = df
df_copy = df.copy()
这样修改df_copy时df不会发生变化
df_copy.loc[0, 'gameId'] = 1234
df.loc[0, 'gameId']
print(df_copy.loc[0, 'gameId'])
1234
过滤行
实际df['blueWins'] > 0返回的是一个true/false的列
df[df['blueWins'] > 0]
# df['blueWins'] > 0
| gameId | blueWins | blueWardsPlaced | blueWardsDestroyed | blueFirstBlood | blueKills | blueDeaths | blueAssists | blueEliteMonsters | blueDragons | ... | redTowersDestroyed | redTotalGold | redAvgLevel | redTotalExperience | redTotalMinionsKilled | redTotalJungleMinionsKilled | redGoldDiff | redExperienceDiff | redCSPerMin | redGoldPerMin | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 5 | 4475365709 | 1 | 18 | 0 | 0 | 5 | 3 | 6 | 1 | 1 | ... | 0 | 15201 | 7.0 | 18060 | 221 | 59 | -698 | -101 | 22.1 | 1520.1 |
| 6 | 4493010632 | 1 | 18 | 3 | 1 | 7 | 6 | 7 | 1 | 1 | ... | 0 | 14463 | 6.4 | 15404 | 164 | 35 | -2411 | -1563 | 16.4 | 1446.3 |
| 9 | 4509433346 | 1 | 13 | 1 | 1 | 4 | 5 | 5 | 1 | 1 | ... | 0 | 16605 | 6.8 | 18379 | 247 | 43 | 1548 | 1574 | 24.7 | 1660.5 |
| 12 | 4515594785 | 1 | 18 | 1 | 1 | 7 | 1 | 11 | 1 | 1 | ... | 0 | 14591 | 6.8 | 17443 | 240 | 50 | -3274 | -1659 | 24.0 | 1459.1 |
| 14 | 4516505202 | 1 | 15 | 3 | 1 | 4 | 4 | 4 | 0 | 0 | ... | 0 | 16192 | 7.0 | 18083 | 242 | 48 | 470 | 187 | 24.2 | 1619.2 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 9872 | 4527650398 | 1 | 12 | 0 | 1 | 7 | 7 | 9 | 0 | 0 | ... | 0 | 16399 | 7.0 | 18001 | 216 | 58 | -756 | -1 | 21.6 | 1639.9 |
| 9873 | 4527878058 | 1 | 18 | 2 | 1 | 12 | 6 | 13 | 0 | 0 | ... | 0 | 15934 | 6.6 | 17027 | 197 | 38 | -2639 | -2364 | 19.7 | 1593.4 |
| 9874 | 4527873286 | 1 | 17 | 2 | 1 | 7 | 4 | 5 | 1 | 1 | ... | 0 | 15246 | 6.8 | 16498 | 229 | 34 | -2519 | -2469 | 22.9 | 1524.6 |
| 9875 | 4527797466 | 1 | 54 | 0 | 0 | 6 | 4 | 8 | 1 | 1 | ... | 0 | 15456 | 7.0 | 18367 | 206 | 56 | -782 | -888 | 20.6 | 1545.6 |
| 9878 | 4523772935 | 1 | 18 | 0 | 1 | 6 | 6 | 5 | 0 | 0 | ... | 0 | 15339 | 6.8 | 17379 | 201 | 46 | -927 | 58 | 20.1 | 1533.9 |
4930 rows × 40 columns
也可以有更复杂的条件
df[(df['blueWardsPlaced'] > 10) & (df['blueWardsPlaced'].isin([9, 15, 17]))]
| gameId | blueWins | blueWardsPlaced | blueWardsDestroyed | blueFirstBlood | blueKills | blueDeaths | blueAssists | blueEliteMonsters | blueDragons | ... | redTowersDestroyed | redTotalGold | redAvgLevel | redTotalExperience | redTotalMinionsKilled | redTotalJungleMinionsKilled | redGoldDiff | redExperienceDiff | redCSPerMin | redGoldPerMin | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2 | 4521474530 | 0 | 15 | 0 | 0 | 7 | 11 | 4 | 1 | 1 | ... | 0 | 17285 | 6.8 | 17254 | 203 | 28 | 1172 | 1033 | 20.3 | 1728.5 |
| 14 | 4516505202 | 1 | 15 | 3 | 1 | 4 | 4 | 4 | 0 | 0 | ... | 0 | 16192 | 7.0 | 18083 | 242 | 48 | 470 | 187 | 24.2 | 1619.2 |
| 15 | 4482120064 | 0 | 17 | 1 | 0 | 3 | 7 | 3 | 0 | 0 | ... | 0 | 17011 | 7.2 | 18778 | 237 | 51 | 1996 | 1804 | 23.7 | 1701.1 |
| 22 | 4480384157 | 0 | 17 | 2 | 0 | 4 | 6 | 3 | 0 | 0 | ... | 0 | 17027 | 7.0 | 18129 | 231 | 60 | 1254 | 567 | 23.1 | 1702.7 |
| 25 | 4523978853 | 0 | 17 | 1 | 0 | 4 | 8 | 4 | 0 | 0 | ... | 0 | 17887 | 7.0 | 17114 | 221 | 36 | 2472 | 1067 | 22.1 | 1788.7 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 9845 | 4527853173 | 0 | 17 | 0 | 0 | 2 | 8 | 1 | 0 | 0 | ... | 0 | 18122 | 7.2 | 19051 | 243 | 50 | 4676 | 3551 | 24.3 | 1812.2 |
| 9851 | 4527637091 | 0 | 15 | 2 | 0 | 4 | 8 | 2 | 0 | 0 | ... | 0 | 17300 | 7.2 | 18342 | 236 | 55 | 2591 | 1250 | 23.6 | 1730.0 |
| 9853 | 4527865649 | 0 | 17 | 2 | 0 | 7 | 8 | 9 | 1 | 0 | ... | 0 | 17206 | 7.0 | 18580 | 212 | 56 | -188 | 676 | 21.2 | 1720.6 |
| 9855 | 4527433973 | 0 | 15 | 3 | 0 | 2 | 7 | 3 | 0 | 0 | ... | 0 | 16148 | 6.6 | 17538 | 198 | 42 | 880 | -179 | 19.8 | 1614.8 |
| 9874 | 4527873286 | 1 | 17 | 2 | 1 | 7 | 4 | 5 | 1 | 1 | ... | 0 | 15246 | 6.8 | 16498 | 229 | 34 | -2519 | -2469 | 22.9 | 1524.6 |
2205 rows × 40 columns
内置函数
DataFrame内置了一些非常方便的函数，比如算均值mean，最小值min，最大值max等等。 API文档
df[cols].mean()
gameId 4.500084e+09
blueWins 4.990384e-01
blueAvgLevel 6.916004e+00
redAvgLevel 6.925316e+00
dtype: float64
某列Series也是一样有很多内置函数 API文档
print(df['redAvgLevel'].mean())
print(df['redAvgLevel'].min())
print(df['redAvgLevel'].max())
6.925316327563518
4.8
8.2
比如可以快速统计出某一列各个数据出现的次数。
df['blueWins'].value_counts()
0 4949
1 4930
Name: blueWins, dtype: int64
apply函数
apply是个非常强大的函数，可以传入一个函数，这一列中的每个元素都会被这个函数作用，最后返回一个新的列。
df['blueWins'].apply(lambda x: 'win' if x==1 else 'lose')
0 lose
1 lose
2 lose
3 lose
4 lose
...
9874 win
9875 win
9876 lose
9877 lose
9878 win
Name: blueWins, Length: 9879, dtype: object
也可以不用lambda表达式，定义更复杂的函数
def win(x):
return 'win' if x == 1 else 'lose'
df_copy['blueWins'] = df['blueWins'].apply(win)
df_copy['blueWins']
0 lose
1 lose
2 lose
3 lose
4 lose
...
9874 win
9875 win
9876 lose
9877 lose
9878 win
Name: blueWins, Length: 9879, dtype: object
列运算
df['brGoldDiff'] = df['blueTotalGold'] - df['redTotalGold']
print(df['brGoldDiff'])
print(df['brGoldDiff'] == df['blueGoldDiff'])
0 643
1 -2908
2 -1172
3 -1321
4 -1004
...
9874 2519
9875 782
9876 -2416
9877 -839
9878 927
Name: brGoldDiff, Length: 9879, dtype: int64
0 True
1 True
2 True
3 True
4 True
...
9874 True
9875 True
9876 True
9877 True
9878 True
Length: 9879, dtype: bool
type(df['brGoldDiff'] == df['blueGoldDiff'])
pandas.core.series.Series
小例子
案例1中可能会要求大家对列数据进行离散话，这里举个例子，把某列数据离散化成最小值到最大值的k个区间。
（实际上还有pandas还有cut和qcut函数可以帮助离散化，感兴趣的同学可以进一步了解使用）。
print(df['blueTotalGold'])
def min_max(x, max_v, min_v, k):
return (x - min_v) // ((max_v - min_v + 1) // k)
min_v = df['blueTotalGold'].min()
max_v = df['blueTotalGold'].max()
df_copy['blueTotalGold'] = df['blueTotalGold'].apply(lambda x: min_max(x, max_v=max_v, min_v=min_v, k=10))
print(df_copy['blueTotalGold'])
df_copy['blueTotalGold'].value_counts()
0 17210
1 14712
2 16113
3 15157
4 16400
...
9874 17765
9875 16238
9876 15903
9877 14459
9878 16266
Name: blueTotalGold, Length: 9879, dtype: int64
0 4
1 3
2 4
3 3
4 4
..
9874 5
9875 4
9876 3
9877 2
9878 4
Name: blueTotalGold, Length: 9879, dtype: int64
4 3236
3 2753
5 1970
2 879
6 721
7 198
1 67
8 40
9 12
0 2
10 1
Name: blueTotalGold, dtype: int64
其他
其他更多应用和函数可查询API文档(中文) API文档(英文)
Numpy
pandas的DataFrame和列Series都可以直接取出数据为numpy的矩阵
print(type(df.values))
print(df.values)
print(type(df['blueTotalGold'].values))
<class 'numpy.ndarray'>
[[ 4.51915782e+09 0.00000000e+00 2.80000000e+01 ... 1.97000000e+01
1.65670000e+03 6.43000000e+02]
[ 4.52337195e+09 0.00000000e+00 1.20000000e+01 ... 2.40000000e+01
1.76200000e+03 -2.90800000e+03]
[ 4.52147453e+09 0.00000000e+00 1.50000000e+01 ... 2.03000000e+01
1.72850000e+03 -1.17200000e+03]
...
[ 4.52771372e+09 0.00000000e+00 2.30000000e+01 ... 2.61000000e+01
1.83190000e+03 -2.41600000e+03]
[ 4.52762831e+09 0.00000000e+00 1.40000000e+01 ... 2.47000000e+01
1.52980000e+03 -8.39000000e+02]
[ 4.52377294e+09 1.00000000e+00 1.80000000e+01 ... 2.01000000e+01
1.53390000e+03 9.27000000e+02]]
<class 'numpy.ndarray'>
数据是没有列名的，类似一个n维数组
blueTotalGold = df['blueTotalGold'].values
print(blueTotalGold)
print(blueTotalGold.dtype)
[17210 14712 16113 ... 15903 14459 16266]
int64
矩阵运算
比n维数组好的地方是支持矩阵运算，比如如果是python数组，要对每个元素+1，需要做循环
python_list = [[1,2,3,4,5], [11,22,33,44,55]]
python_list = [[i + 1 for i in l] for l in python_list]
print(python_list)
[[2, 3, 4, 5, 6], [12, 23, 34, 45, 56]]
但是numpy方便矩阵运算
numpy_array = np.array([[1,2,3,4,5], [11,22,33,44,55]])
print(numpy_array + 1)
[[ 2 3 4 5 6]
[12 23 34 45 56]]
新建特殊矩阵
print(np.zeros((3, 4)))
print(np.ones((2, 2)))
print(np.empty( (2,3) ) )
print(np.arange( 10, 30, 5 ))
print(np.arange(12).reshape(4,3))
[[0. 0. 0. 0.]
[0. 0. 0. 0.]
[0. 0. 0. 0.]]
[[1. 1.]
[1. 1.]]
[[8.5029e-320 7.2687e-320 7.9609e-320]
[7.8571e-320 7.1437e-320 8.0365e-320]]
[10 15 20 25]
[[ 0 1 2]
[ 3 4 5]
[ 6 7 8]
[ 9 10 11]]
矩阵属性信息
print(df[cols])
arr = df[cols].values
print(arr)
print(arr.shape)
print(arr.ndim)
print(arr.size)
print(arr.dtype)
gameId blueWins blueAvgLevel redAvgLevel
0 4519157822 0 6.6 6.8
1 4523371949 0 6.6 6.8
2 4521474530 0 6.4 6.8
3 4524384067 0 7.0 7.0
4 4436033771 0 7.0 7.0
... ... ... ... ...
9874 4527873286 1 7.2 6.8
9875 4527797466 1 7.2 7.0
9876 4527713716 0 7.0 7.4
9877 4527628313 0 6.6 7.2
9878 4523772935 1 7.0 6.8
[9879 rows x 4 columns]
[[4.51915782e+09 0.00000000e+00 6.60000000e+00 6.80000000e+00]
[4.52337195e+09 0.00000000e+00 6.60000000e+00 6.80000000e+00]
[4.52147453e+09 0.00000000e+00 6.40000000e+00 6.80000000e+00]
...
[4.52771372e+09 0.00000000e+00 7.00000000e+00 7.40000000e+00]
[4.52762831e+09 0.00000000e+00 6.60000000e+00 7.20000000e+00]
[4.52377294e+09 1.00000000e+00 7.00000000e+00 6.80000000e+00]]
(9879, 4)
2
39516
float64
元素访问
访问第2行
arr[1, :]
array([4.52337195e+09, 0.00000000e+00, 6.60000000e+00, 6.80000000e+00])
访问第3列
arr[:, 2]
array([6.6, 6.6, 6.4, ..., 7. , 6.6, 7. ])
访问前3行，3-最后一列
arr[:3, 2:]
array([[6.6, 6.8],
[6.6, 6.8],
[6.4, 6.8]])
数学运算
arr[:3, 2:] + 1
array([[7.6, 7.8],
[7.6, 7.8],
[7.4, 7.8]])
(arr[:3, 2:] - 1) * 10
array([[56., 58.],
[56., 58.],
[54., 58.]])
arr[:3, 2:] ** 2
array([[43.56, 46.24],
[43.56, 46.24],
[40.96, 46.24]])
arr[:3, 2:].sum()
40.0
arr[:3, 2:].mean()
6.666666666666667
arr[:3, 2:].max()
arr[:3, 2:].min()
6.4
np.sum(arr[:3, 2:])
np.sin(arr[:3, 2:])
np.cos(arr[:3, 2:])
array([[0.95023259, 0.86939749],
[0.95023259, 0.86939749],
[0.99318492, 0.86939749]])
*是元素乘，@是矩阵乘
A = np.array( [[1,1],
[0,1]] )
B = np.array( [[2,0],
[3,4]] )
A * B
array([[2, 0],
[0, 4]])
A @ B
array([[5, 4],
[3, 4]])
A + B
array([[3, 1],
[3, 5]])
A - B
array([[-1, 1],
[-3, -3]])
按元素相除，除0会抛出warning，返回无穷inf
A / B
/home/shisy13/anaconda3/envs/conda3/lib/python3.7/site-packages/ipykernel_launcher.py:1: RuntimeWarning: divide by zero encountered in true_divide
"""Entry point for launching an IPython kernel.
array([[0.5 , inf],
[0. , 0.25]])
数学运算矩阵的大小必须要合法
# A * np.array([1, 2, 3])
A * np.array([1, 2])
array([[1, 2],
[0, 2]])
拷贝
和pandsa一样，一般等号负值是引用，copy之后才是新的对象复制
A_copy = A.copy()
A_copy[0,0] = 2
print(A)
print(A_copy)
[[1 1]
[0 1]]
[[2 1]
[0 1]]
小例子
假设要对平均等级做离散化，用类似之前pandas里的实现方式，最大最小值之间以0.2分割区间
blueAvgLevel = df['blueAvgLevel'].values
print(blueAvgLevel)
[6.6 6.6 6.4 ... 7. 6.6 7. ]
print(blueAvgLevel.min(), blueAvgLevel.max())
4.6 8.0
float运算后还是float
(blueAvgLevel - 4.6)/0.1
array([20., 20., 18., ..., 24., 20., 24.])
np.around()可以四舍五入，astype可以变换数据类型
print((blueAvgLevel - 4.6)/0.2)
blueAvgLevel_new = np.around((blueAvgLevel - 4.6)/0.2).astype(int)
print(blueAvgLevel_new)
[10. 10. 9. ... 12. 10. 12.]
[10 10 9 ... 12 10 12]
可以把array赋值给pandas的某一列
df['blueAvgLevel'] = blueAvgLevel_new
print(df['blueAvgLevel'])
df[cols]
del df
del arr
0 10
1 10
2 9
3 12
4 12
..
9874 13
9875 13
9876 12
9877 10
9878 12
Name: blueAvgLevel, Length: 9879, dtype: int64