Question

hd.loc[(hd['ap_hi'] <= 120) & (hd['ap_lo'] < 80), ['ap_hi','ap_lo']] = 'normal'
hd.loc[(hd['ap_hi'] > 120) & (hd['ap_hi'] <= 129) & (hd['ap_lo'] < 80), ['ap_hi','ap_lo']] = 'elevated'
hd.loc[(hd['ap_hi'] > 130) & (hd['ap_hi'] <= 139) | (hd['ap_lo'] >= 80) & (hd['ap_lo'] < 89), ['ap_hi','ap_lo']] = 'high blood pressure 1'
hd.loc[(hd['ap_hi'] > 140) & (hd['ap_hi'] <= 179) | (hd['ap_lo'] > 90) & (hd['ap_lo'] <119 ), ['ap_hi','ap_lo']] = 'high blood pressure 2'
hd.loc[(hd['ap_hi'] > 180) | (hd['ap_lo'] > 120) , ['ap_hi','ap_lo']] = 'hypertensive crisis'

运行此代码时，第二行出现错误'>' not supported between instances of 'str' and 'int'。我不知道是什么引起了错误。预先感谢。

Answer 1

您正在将字符串值放在包含整数的列中。而不是这样做，为字符串创建新的列。在这里，我正在创建一个新列'bp_level'：

hd.loc[(hd['ap_hi'] <= 120) & (hd['ap_lo'] < 80), 'bp_level'] = 'normal'
hd.loc[(hd['ap_hi'] > 120) & (hd['ap_hi'] <= 129) & (hd['ap_lo'] < 80), 'bp_level'] = 'elevated'
hd.loc[(hd['ap_hi'] > 130) & (hd['ap_hi'] <= 139) | (hd['ap_lo'] >= 80) & (hd['ap_lo'] < 89), 'bp_level'] = 'high blood pressure 1'
hd.loc[(hd['ap_hi'] > 140) & (hd['ap_hi'] <= 179) | (hd['ap_lo'] > 90) & (hd['ap_lo'] <119 ), 'bp_level'] = 'high blood pressure 2'
hd.loc[(hd['ap_hi'] > 180) | (hd['ap_lo'] > 120) , 'bp_level'] = 'hypertensive crisis'

如果要覆盖这些列，请在完成所有比较之后执行此操作：

hd.loc[:,['ap_hi', 'ap_lo']] = hd['bp_level']

这是一个更简单，有效的示例（已通过Python 3.8和pandas 1.0.5测试）：

import pandas as pd
df = pd.DataFrame({'A':range(10)})
df.loc[(df['A'] < 3), 'B'] = '<3'
df.loc[(df['A'] < 6) & (df['A'] >= 3), 'B'] = '3 to <6'
df.loc[(df['A'] >= 6), 'B'] = '6+'

print(df)

产生：

   A        B
0  0       <3
1  1       <3
2  2       <3
3  3  3 to <6
4  4  3 to <6
5  5  3 to <6
6  6       6+
7  7       6+
8  8       6+
9  9       6+

'str'和'int'错误的实例之间不支持'>'

1 个答案: