pandas Python - 根据数据帧列中的数据将对象数据类型转换为整数、字符串或浮点数
声明：本页面是StackOverFlow热门问题的中英对照翻译，遵循CC BY-SA 4.0协议，如果您需要使用它，必须同样遵循CC BY-SA许可，注明原文地址和作者信息，同时你必须将它归于原作者（不是我）：StackOverFlow
原文地址: http://stackoverflow.com/questions/47785101/
Warning: these are provided under cc-by-sa 4.0 license. You are free to use/share it, But you must attribute it to the original authors (not me):
StackOverFlow
Python - convert object data type to integer, string or float based on data in dataframe column
提问by vbala2014
I have the following issue. I have a dataframe which has various types of columns (int, float, string, etc) - but since they were imported into python using a .csv file all columns are showing as objectdata type. Example is below:
我有以下问题。我有一个数据框，其中包含各种类型的列（int、float、string 等） - 但由于它们是使用 .csv 文件导入到 python 中的，因此所有列都显示为object数据类型。示例如下：
print df_centers
output:
输出：
center name ID state activity type cost usage
Bay area recreational facility 10019 LA swimming 0.5%
Ith area recreational facility 10020 NY basketball 0 100%
all of these columns have data type = object, i am trying to convert the object data types to their relevant and more meaningful data types. Example below:
所有这些列都具有数据类型 = 对象，我正在尝试将对象数据类型转换为其相关且更有意义的数据类型。下面的例子：
df_centers['cost'] = df_centers['cost'].astype('int')
The reason i am trying to do an INT conversion on the cost field is because i have to do some analysis later on it but python returns with the following error message:
我尝试在成本字段上进行 INT 转换的原因是因为我稍后必须对其进行一些分析，但是 python 返回并显示以下错误消息：
ValueError: invalid literal for long() with base 10: ''
i have also tried converting usage to 0.5% to float with the below and it returns the error message:
我还尝试将使用率转换为 0.5% 以浮动以下内容并返回错误消息：
df_centers['usage'] = df_centers['usage'].astype('float')
output i get is:
我得到的输出是：
invalid literal for float(): 100%
Any suggestions on how i can get this data type conversion done from object to a more relevant type?
关于如何将这种数据类型从对象转换为更相关类型的任何建议？
回答by Gary02127
Usually in CSV files, if it's not just comma-delimited but rather an Excel file, etc, the "object" has a type and value which may help you decipher what's what.
通常在 CSV 文件中，如果它不仅仅是逗号分隔而是 Excel 文件等，则“对象”具有一个类型和值，可以帮助您破译什么是什么。
In the interim, to convert money values to numbers, strip off the leading '$' and convert to float. For the percentages, strip off the %, convert the number to float, then divide it by 100.
在此期间，要将货币值转换为数字，请去除前导“$”并转换为浮点数。对于百分比，去掉 %，将数字转换为浮点数，然后除以 100。
So, this:
所以这：
df_centers['cost'] = df_centers['cost'].astype('int')
df_centers['usage'] = df_centers['usage'].astype('float')
should be:
应该：
df_centers['cost'] = df_centers['cost'].str.lstrip('$').astype('int')
# ^^^^^^^^^^^^^^^^
df_centers['usage'] = df_centers['usage'].str.rstrip('%').astype('float') / 100.0
# ^^^^^^^^^^^^^^^^
