将会计格式的 pandas 系列转换为数字系列？ [英] Convert a Pandas Series in Accounting Format to a Numeric Series?

问题描述

数值的会计格式通常使用货币字符，通常使用括号来表示负值。零也可以表示为 - 或 $ - 。当这样的系列导入到Pandas DataFrame中时，它是一个对象类型。我需要将其转换为数字类型并正确解析负值。

An accounting format for numeric values usually uses a currency character, and often uses parentheses to represent negative values. Zero may also be represented as a - or $-. When such a series is imported into a Pandas DataFrame it is an object type. I need to convert it to a numeric type and parse the negative values correctly.

这里有一个例子：

import numpy as np
import pandas as pd
from pandas import Series, DataFrame
df = pd.DataFrame({'A':['123.4', '234.5', '345.5', '456.7'],
                   'B':['$123.4', '$234.5', '$345.5', '$456.7'],
                   'C':['($123.4)', '$234.5', '($345.5)', '$456.7'],
                   'D':['$123.4', '($234.5)', '$-', '$456.7']})

系列A很容易转换，例如

Series A is easy to convert e.g.

df['A'] = df['A'].astype(float) 

系列B需要删除 $ 标志，之后它是直接的。

Series B required the removal of the $ sign, after which it is then straightforward.

然后是C和D系列。它们包含括号（即否定）值为D，零包含 $ - 。如何正确地将这些系列解析成数字系列/数据框？

Then comes series C and D. They contain parentheses (i.e. negative) values and D contains $- for zero. How can I correctly parse theses series into numeric series / dataframe?

推荐答案

我会使用熊猫替换函数来替换$和）没有，替换为0，然后最后替换（by - 。然后你可以做 df = astype（float）它应该工作。

I'd use the Pandas replace function to replace $ and ) by nothing, replace - by 0, and then finally replace ( by -. Then you can do df=astype(float) and it should work.

这篇关于将会计格式的 pandas 系列转换为数字系列？的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！