CODEWITHSUNDEEP
pythonpandasdata-science
Emm
Emm

Reputation: 2507

Creating dataframe from multiple pandas series

Trying to create a dataframe from multiple pandas series.

Here is my code

series = [pd.Series([nine, ten, eleven, twelve,thirteen], index=('link','post','shared','status','timeline'))]
my_names = [2009,2010,2011,2012,2013,2014,2015,2016,2017,2018]
my_series = [pd.Series([nine, ten, eleven, twelve,thirteen], index=('link','post','shared','status','timeline'), name=n) for n in my_names]

Here is a sample of the data series I am interacting with:

#nine
title
link       6.060606
post       8.080808
status    85.858586
dtype: float64

#ten
title
link         1.666667
post        15.833333
shared       1.666667
status      71.666667
timeline     9.166667
dtype: float64

#eleven
title
link        23.885350
shared       0.318471
status      59.872611
timeline    15.923567
dtype: float64

#twelve
title
link        18.660287
shared       3.588517
status      65.550239
timeline    12.200957
dtype: float64

The desired outcome is having 2009-2018 as the column titles, link, shared, status, timeline post as the row indexes and populating each row with the values from each of the series (nine =2009, ten=2010).

This is a sample of my current outcome:

sample

Upvotes: 1

Views: 2318

Answers (1)

jezrael
jezrael

Reputation: 863761

I believe you need concat with parameter keys for new column names:

ser = [nine, ten, eleven, twelve,thirteen]
my_names = [2009,2010,2011,2012,2013]

df = pd.concat(ser, axis=1, keys=my_names)

If input are lists:

nine = [4,5,6]
ten = [9,0,3]
eleven = [7,3,0]

L = [nine, ten, eleven]
my_names = [2009,2010,2011]

df = pd.DataFrame({k:v for k, v in zip(my_names, L)})
print (df)
   2009  2010  2011
0     4     9     7
1     5     0     3
2     6     3     0

Your solution return list of Series, so only need concat:

df = pd.concat(my_series, axis=1)

Or:

df = pd.DataFrame(my_series).T

Upvotes: 1

Related Questions