Python3.4：字典中的DataFrame

本文介绍了Python3.4：字典中的DataFrame的处理方法，对大家解决问题具有一定的参考价值，需要的朋友们下面随着小编来一起学习吧！ 问题描述

以下代码旨在填充熊猫DataFrame。

The following code is meant to populate a pandas DataFrame. Unfortunately the output is coming out weird.

output = {'Revenue' : [1456216, 549514, 489461], 'Cost of Revenue' : [1565486, 498464, 156131], 'Gross Profit' : [456465, 565165, 651613] ... ... ... } years = ['year1', 'year2', 'year3'] df = pd.DataFrame(output.values(), index=output.keys(), columns=years) print(df)

字典 output 具有列表值。我想使用output.keys（）作为DataFrame索引，使用列的列出年份，并使用output.values（）作为数据框架内的数据。最终，我希望输出如下

The dictionary output has list values. I want to use output.keys() as the DataFrame index, the list years for the columns and output.values() as the data within the data frame. Ultimately I want the ouput to be as follows

Year 1 Year 2 Year 3 Revenue 1456216 549514 489461 Cost of Revenue 1565486 498464 156131 Gross Profit 456465 565165 651613 ... ... ... ... ... ... ... ...

我该怎么做？

year1 \ Total Revenue ([2218767, 1528545, 972309], [293797, 202908, ... Cost of Revenue ([2218767, 1528545, 972309], [293797, 202908, ... Gross Profit ([2218767, 1528545, 972309], [293797, 202908, ... Research Development ([2218767, 1528545, 972309], [293797, 202908, ... Selling General and Administrative ([2218767, 1528545, 972309], [293797, 202908, ... Non Recurring ([2218767, 1528545, 972309], [293797, 202908, ... Others ([2218767, 1528545, 972309], [293797, 202908, ... Total Operating Expenses ([2218767, 1528545, 972309], [293797, 202908, ... Operating Income or Loss ([2218767, 1528545, 972309], [293797, 202908, ... Total Other Income/Expenses Net ([2218767, 1528545, 972309], [293797, 202908, ... Earnings Before Interest And Taxes ([2218767, 1528545, 972309], [293797, 202908, ... Interest Expense ([2218767, 1528545, 972309], [293797, 202908, ... Income Before Tax ([2218767, 1528545, 972309], [293797, 202908, ... Income Tax Expense ([2218767, 1528545, 972309], [293797, 202908, ... Minority Interest ([2218767, 1528545, 972309], [293797, 202908, ... Net Income From Continuing Ops ([2218767, 1528545, 972309], [293797, 202908, ... Discontinued Operations ([2218767, 1528545, 972309], [293797, 202908, ... Extraordinary Items ([2218767, 1528545, 972309], [293797, 202908, ... Effect Of Accounting Changes ([2218767, 1528545, 972309], [293797, 202908, ... Other Items ([2218767, 1528545, 972309], [293797, 202908, ... Net Income ([2218767, 1528545, 972309], [293797, 202908, ... Preferred Stock And Other Adjustments ([2218767, 1528545, 972309], [293797, 202908, ... Net Income Applicable To Common Shares ([2218767, 1528545, 972309], [293797, 202908, ... year2 \ Total Revenue ([2218767, 1528545, 972309], [293797, 202908, ... Cost of Revenue ([2218767, 1528545, 972309], [293797, 202908, ... Gross Profit ([2218767, 1528545, 972309], [293797, 202908, ... Research Development ([2218767, 1528545, 972309], [293797, 202908, ... Selling General and Administrative ([2218767, 1528545, 972309], [293797, 202908, ... Non Recurring ([2218767, 1528545, 972309], [293797, 202908, ... Others ([2218767, 1528545, 972309], [293797, 202908, ... Total Operating Expenses ([2218767, 1528545, 972309], [293797, 202908, ... Operating Income or Loss ([2218767, 1528545, 972309], [293797, 202908, ... Total Other Income/Expenses Net ([2218767, 1528545, 972309], [293797, 202908, ... Earnings Before Interest And Taxes ([2218767, 1528545, 972309], [293797, 202908, ... Interest Expense ([2218767, 1528545, 972309], [293797, 202908, ... Income Before Tax ([2218767, 1528545, 972309], [293797, 202908, ... Income Tax Expense ([2218767, 1528545, 972309], [293797, 202908, ... Minority Interest ([2218767, 1528545, 972309], [293797, 202908, ... Net Income From Continuing Ops ([2218767, 1528545, 972309], [293797, 202908, ... Discontinued Operations ([2218767, 1528545, 972309], [293797, 202908, ... Extraordinary Items ([2218767, 1528545, 972309], [293797, 202908, ... Effect Of Accounting Changes ([2218767, 1528545, 972309], [293797, 202908, ... Other Items ([2218767, 1528545, 972309], [293797, 202908, ... Net Income ([2218767, 1528545, 972309], [293797, 202908, ... Preferred Stock And Other Adjustments ([2218767, 1528545, 972309], [293797, 202908, ... Net Income Applicable To Common Shares ([2218767, 1528545, 972309], [293797, 202908, ... year3 You get the idea OrderedDict([('Total Revenue', [2218767, 1528545, 972309]), ('Cost of Revenue', [293797, 202908, 125521]), ('Gross Profit', [1924970, 1325637, 846788]), ('Research Development', [536184, 395643, 257179]), ('Selling General and Administrative', [1115705, 747666, 452898]), ('Non Recurring', ['0', '0', '0']), ('Others', [236946, 134516, 79849]), ('Total Operating Expenses', ['0', '0', '0']), ('Operating Income or Loss', [36135, 47812, 56862]), ('Total Other Income/Expenses Net', [-4930, 1416, 252]), ('Earnings Before Interest And Taxes', [31205, 49228, 57114]), ('Interest Expense', ['0', '0', '0']), ('Income Before Tax', [31205, 49228, 57114]), ('Income Tax Expense', [46525, 22459, 35504]), ('Minority Interest', [-427, '0', '0']), ('Net Income From Continuing Ops', [-15747, 26769, 21610]), ('Discontinued Operations', ['0', '0', '0']), ('Extraordinary Items', ['0', '0', '0']), ('Effect Of Accounting Changes', ['0', '0', '0']), ('Other Items', ['0', '0', '0']), ('Net Income', [-15747, 26769, 21610]), ('Preferred Stock And Other Adjustments', ['0', '0', '0']), ('Net Income Applicable To Common Shares', [-15747, 26769, 21610])])

推荐答案

您可以传递 list（output.values（））而不是传递 output.values（）：

Instead of passing output.values(), you could pass list(output.values()):

>>> pd.DataFrame(list(output.values()), index=output.keys(), columns=years) year1 year2 year3 Cost of Revenue 1565486 498464 156131 Gross Profit 456465 565165 651613 Revenue 1456216 549514 489461

您也可以写

>>> df = pd.DataFrame.from_dict(output, orient='index') >>> df.columns = years >>> df year1 year2 year3 Cost of Revenue 1565486 498464 156131 Gross Profit 456465 565165 651613 Revenue 1456216 549514 489461

要求DataFrame构造函数执行很多操作，有时我认为将其分解更简单。

The DataFrame constructor is asked to do a lot of things, and sometimes I think it's simpler to break it up.

关于为什么发生这种情况，构造函数中有很多特殊的类型，并且该分支如下：

As for why this is happening, there's a lot of special-casing on type in the constructor, and this branch:

elif isinstance(data, (list, types.GeneratorType)):

不会选择 dict_values 对象，因为它既不是列表也不是生成器。

isn't picking up the dict_values object because it's neither a list nor a generator.