Renaming columns of dataframe list in Pandas

Question

I have a list with lots of dataframes

col = ['open', 'high', 'low', 'close'] index = [1, 2, 3, 4] df1 = pd.DataFrame(columns=col, index=index) df2 = pd.DataFrame(columns=col, index=index) df3 = pd.DataFrame(columns=col, index=index) dflist = [df1, df2, df3]

I need to rename all the columns of all the dataframes in the list. I need to add the name of each dataframe to the name of each column. I tried to do it with a for loop.

for key in dflist: key.rename(columns=lambda x: key+x)

Obviously, this is not working. The desired output would be:

In [1]: df1.columns.tolist() Out [2]: ['df1open', 'df1high', 'df1low', 'df1close'] In [3]: df2.columns.tolist() Out [4]: ['df2open', 'df2high', 'df2low', 'df2close'] In [5]: df3.columns.tolist() Out [6]: ['df3open', 'df3high', 'df3low', 'df3close']

Thanks for your help.

YS-L · Accepted Answer · 2014-09-02 00:25:14Z

You want to use a dict instead of a list to store the DataFrames, if you need to somehow access their "names" and manipulate them programmatically (think when you have thousands of them). Also note the use of the inplace argument, which is common in pandas:

import pandas as pd col = ['open', 'high', 'low', 'close'] index = [1, 2, 3, 4] df_all = {'df1': pd.DataFrame(columns=col, index=index), 'df2': pd.DataFrame(columns=col, index=index), 'df3': pd.DataFrame(columns=col, index=index)} for key, df in df_all.iteritems(): df.rename(columns=lambda x: key+x, inplace=True) print df_all['df1'].columns.tolist()

Output:

['df1open', 'df1high', 'df1low', 'df1close']

chrisaycock · Accepted Answer · 2014-09-02 00:20:45Z

There are a couple of issues here. Firstly, dflist is the list of DataFrames, as opposed to the names of those DataFrames. So df1 is not the same as "df1", which means that key + x isn't a string concatenation.

Secondly, the rename() function returns a new DataFrame. So you have to pass the inplace=True parameter to overwrite the existing column names.

Try this instead:

dflist = ['df1', 'df2', 'df3'] for key in dflist: df = eval(key) df.rename(columns=lambda x: key+x, inplace=True)

Collectives™ on Stack Overflow

Renaming columns of dataframe list in Pandas

2 Answers 2

Comments

Comments

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Related