如何选择数据
作者：向阳逐梦
- 2022-10-31 四川
本文字数：4901 字
阅读完需：约 16 分钟
1. Series
1.1 索引
Series 对象索引的工作原理和 ndarray 对象索引非常类似，不同的一点是，在对 Series 对象进行索引时，我们不但可以使用整数还可以使用 Series 对象本身的索引，举几个例子。单个元素：
import pandas as pd
my_series = pd.Series([4, -7, 6, -5, 3, 2], index=["a", "b", "c", "d", "e", "f"])
print(my_series[3])print(my_series['d'])复制代码
在上面的代码中，my_series[3] 和 my_series['d'] 访问的是同一个元素，一个使用的是整数 3，一个使用的是 Series 对象本身的索引 d。
多个元素：
import pandas as pd
my_series = pd.Series([4, -7, 6, -5, 3, 2], index=["a", "b", "c", "d", "e", "f"])
print(my_series[[1, 3, 4]])print(my_series[['b', 'd', 'e']])复制代码
切片：
import pandas as pd
my_series = pd.Series([4, -7, 6, -5, 3, 2], index=["a", "b", "c", "d", "e", "f"])
print(my_series[0:5])print(my_series['a':'e'])复制代码
使用整数进行切片和使用 Series 对象的索引进行切片不同的地方是，使用整数切片不包含最末的那个元素，而使用对象的索引进行切片时是包含最末的那个元素的。
布尔索引：
import pandas as pd
my_series = pd.Series([4, -7, 6, -5, 3, 2], index=["a", "b", "c", "d", "e", "f"])
print(my_series[my_series > my_series.median()])复制代码
上面代码的输出结果是大于平均值的元素。
1.2 像字典一样选择数据
import pandas as pdimport numpy as np
my_series = pd.Series([4, -7, 6, -5, 3, 2], index=["a", "b", "c", "d", "e", "f"])
print(my_series['b'])print(my_series.get('b', np.NaN))复制代码
判断 key 是否在 Series 里：
import pandas as pdimport numpy as np
my_series = pd.Series([4, -7, 6, -5, 3, 2], index=["a", "b", "c", "d", "e", "f"])
print('e' in my_series)print('g' in my_series)复制代码
1.3 loc 或 iloc
loc 和 iloc 的区别是，loc 使用的是轴标签，iloc 使用的是整数。
单个元素：
import pandas as pdimport numpy as np
my_series = pd.Series([4, -7, 6, -5, 3, 2], index=["a", "b", "c", "d", "e", "f"])
print(my_series.iloc[1])print(my_series.loc['b'])复制代码
多个元素：
import pandas as pdimport numpy as np
my_series = pd.Series([4, -7, 6, -5, 3, 2], index=["a", "b", "c", "d", "e", "f"])
print(my_series.iloc[[1, 3]])print(my_series.loc[['b', 'd']])复制代码
切片：
import pandas as pdimport numpy as np
my_series = pd.Series([4, -7, 6, -5, 3, 2], index=["a", "b", "c", "d", "e", "f"])
print(my_series.iloc[1:5])print(my_series.loc['b': 'e'])复制代码
使用整数进行切片和使用 Series 对象的索引进行切片不同的地方是，使用整数切片不包含最末的那个元素，而使用对象的索引进行切片时是包含最末的那个元素的。
2. DataFrame
2.1 索引
使用索引的方法来访问 DataFrame 时，得到的是一列或多列。单列的访问：
import pandas as pd
d = { "Open": pd.Series([136, 137, 140, 143, 141, 142], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "High": pd.Series([137, 140, 143, 144, 144, 145], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "Low": pd.Series([135, 137, 140, 142, 140, 142], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "Close": pd.Series([137, 139, 142, 144, 143, 145], index = ['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09'])}
df = pd.DataFrame(d)print(df['Open'])复制代码
上面的单列访问我们还可以使用另外一种方法，例如：
import pandas as pd
d = { "Open": pd.Series([136, 137, 140, 143, 141, 142], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "High": pd.Series([137, 140, 143, 144, 144, 145], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "Low": pd.Series([135, 137, 140, 142, 140, 142], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "Close": pd.Series([137, 139, 142, 144, 143, 145], index = ['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09'])}
df = pd.DataFrame(d)
print(df.Open)复制代码
上面代码采用的方法是类似于属性的访问方法。多列的访问：
import pandas as pd
d = { "Open": pd.Series([136, 137, 140, 143, 141, 142], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "High": pd.Series([137, 140, 143, 144, 144, 145], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "Low": pd.Series([135, 137, 140, 142, 140, 142], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "Close": pd.Series([137, 139, 142, 144, 143, 145], index = ['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09'])}
df = pd.DataFrame(d)print(df[['Open', 'Close']])复制代码
布尔索引：
import pandas as pd
d = { "Open": pd.Series([136, 137, 140, 143, 141, 142], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "High": pd.Series([137, 140, 143, 144, 144, 145], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "Low": pd.Series([135, 137, 140, 142, 140, 142], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "Close": pd.Series([137, 139, 142, 144, 143, 145], index = ['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09'])}
df = pd.DataFrame(d)print(df[df['Open'] > 140])复制代码
上面代码中，Open 列的值大于 140 的为后面三行，所以最后得到整个 DataFrame 的后面三行。使用布尔索引，我们还可以得到整个 DataFrame 中大于 140 的元素，例如：
import pandas as pd
d = { "Open": pd.Series([136, 137, 140, 143, 141, 142], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "High": pd.Series([137, 140, 143, 144, 144, 145], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "Low": pd.Series([135, 137, 140, 142, 140, 142], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "Close": pd.Series([137, 139, 142, 144, 143, 145], index = ['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09'])}
df = pd.DataFrame(d)print(df[df > 140])复制代码
上面的代码得到了整个 DataFrame 中值大于 140 的元素，不大于 140 的元素用 NaN 填充。
2.2 切片
import pandas as pd
d = { "Open": pd.Series([136, 137, 140, 143, 141, 142], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "High": pd.Series([137, 140, 143, 144, 144, 145], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "Low": pd.Series([135, 137, 140, 142, 140, 142], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "Close": pd.Series([137, 139, 142, 144, 143, 145], index = ['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09'])}
df = pd.DataFrame(d)
print(df[:3])print(df[::2])print(df[::-1])print(df[1:2])复制代码
上面的切片都是针对行来的。
2.3 loc 或 iloc
2.3.1 单行或单列的访问
import pandas as pd
d = { "Open": pd.Series([136, 137, 140, 143, 141, 142], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "High": pd.Series([137, 140, 143, 144, 144, 145], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "Low": pd.Series([135, 137, 140, 142, 140, 142], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "Close": pd.Series([137, 139, 142, 144, 143, 145], index = ['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09'])}
df = pd.DataFrame(d)print(df.iloc[0])print(df.loc['2021-07-01'])print(df.iloc[:, 0])print(df.loc[:, 'Open'])复制代码
2.3.2 多行或多列的访问
import pandas as pd
d = { "Open": pd.Series([136, 137, 140, 143, 141, 142], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "High": pd.Series([137, 140, 143, 144, 144, 145], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "Low": pd.Series([135, 137, 140, 142, 140, 142], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "Close": pd.Series([137, 139, 142, 144, 143, 145], index = ['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09'])}
df = pd.DataFrame(d)print(df.iloc[[0, 1, 2]])print(df.iloc[:, [0, 1, 2]])print(df.loc[:, ['Open', 'High', 'Low']])复制代码
2.3.3 位于行列交叉的元素
import pandas as pd
d = { "Open": pd.Series([136, 137, 140, 143, 141, 142], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "High": pd.Series([137, 140, 143, 144, 144, 145], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "Low": pd.Series([135, 137, 140, 142, 140, 142], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "Close": pd.Series([137, 139, 142, 144, 143, 145], index = ['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09'])}
df = pd.DataFrame(d)
print(df.loc[['2021-07-01', '2021-07-02'], ['Open', 'High']])print(df.iloc[[0, 1], [0, 1]])复制代码
切片
import pandas as pd
d = { "Open": pd.Series([136, 137, 140, 143, 141, 142], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "High": pd.Series([137, 140, 143, 144, 144, 145], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "Low": pd.Series([135, 137, 140, 142, 140, 142], index=['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09']), "Close": pd.Series([137, 139, 142, 144, 143, 145], index = ['2021-07-01', '2021-07-02', '2021-07-06', '2021-07-07', '2021-07-08', '2021-07-09'])}
df = pd.DataFrame(d)
print(df.loc['2021-07-01':'2021-07-07', 'Open': 'Low'])print(df.iloc[0:4, 0:3])复制代码
划线
评论
复制
发布于: 刚刚阅读数: 3
版权声明: 本文为 InfoQ 作者【向阳逐梦】的原创文章。
原文链接:【http://xie.infoq.cn/article/edb5485333a7c6e4109f5eb04】。文章转载请联系作者。
向阳逐梦
关注
人生享受编程，编程造就人生！ 2022-06-01 加入
某公司芯片测试工程师，嵌入式开发工程师，InfoQ签约作者，阿里云星级博主，华为云·云享专家。座右铭：向着太阳，追逐梦想！
评论