I have the following list:
index_list = ['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h']
I would like to use it to define an index for the following DataFrame:
df = pd.DataFrame(index= [0,1,4,7])
The DataFrames index corresponds to an entry in the list. The end result should look like:
df_target = pd.DataFrame(index= ['a','b','e','h'])
I have tried a number of built in functions for pandas, but no luck.
Use Index.map with dictonary created by enumerate:
target_df = pd.DataFrame(index=df.index.map(dict(enumerate(index_list))))
print (target_df)
Empty DataFrame
Columns: []
Index: [a, b, e, h]
If need change index in existing DataFrame:
df = pd.DataFrame({'a':range(4)}, index= [0,1,4,7])
print (df)
a
0 0
1 1
4 2
7 3
df = df.rename(dict(enumerate(index_list)))
print (df)
a
a 0
b 1
e 2
h 3
like this:
print(pd.DataFrame(index=[index_list[i] for i in [0,1,4,7]]))
Output:
Empty DataFrame
Columns: []
Index: [a, b, e, h]
You can pass a Series object to Index.map method.
df.index.map(pd.Series(index_list))
# Index(['a', 'b', 'e', 'h'], dtype='object')
Or rename the Dataframe index directly
df.rename(index=pd.Series(index_list))
# Empty DataFrame
# Columns: []
# Index: [a, b, e, h]
