CODEWITHSUNDEEP
pythonpandasdataframe
bloodynri
bloodynri

Reputation: 603

Convert lists to a diagonal matrix DataFrame

I get returned a list of tuples in this format:

[date1, date2, date3]
[ticker1, ticker2, ticker3]
[value1, value2, value3]

I want to convert it into a pandas dataframe where the index are the dates, columns are the tickers and the data are the values.

Upvotes: 1

Views: 302

Answers (3)

BENY
BENY

Reputation: 323226

Using pivot 

df=pd.DataFrame(lst).T
df.pivot(*df)

1     ticker1 ticker2 ticker3
0                            
date1  value1    None    None
date2    None  value2    None
date3    None    None  value3

Upvotes: 3

Simeon Ikudabo
Simeon Ikudabo

Reputation: 2190

values = [(1, 0, 1), (2, 9, 4), (3, 10, 4), (9, 22, 30)]

data_frame = pd.DataFrame(values)
print(data_frame)

I would need to see your specific code, because what you outlined appears to be a list of lists, not a list of tuples. But in theory, if you had a list of tuples, this is how you would create a DataFrame out of it. Here is your output:

   0   1   2
0  1   0   1
1  2   9   4
2  3  10   4
3  9  22  30

Upvotes: 1

piRSquared
piRSquared

Reputation: 294228

Series and unstack

lst = [['date1', 'date2', 'date3'],
       ['ticker1', 'ticker2', 'ticker3'],
       ['value1', 'value2', 'value3']]

pd.Series(lst[-1], lst[:2]).unstack()

      ticker1 ticker2 ticker3
date1  value1    None    None
date2    None  value2    None
date3    None    None  value3

defaultdict

from collections import defaultdict

lst = [['date1', 'date2', 'date3'],
       ['ticker1', 'ticker2', 'ticker3'],
       ['value1', 'value2', 'value3']]

dd = defaultdict(dict)

for d, t, v in zip(*lst):
    dd[t][d] = v

pd.DataFrame(dd)

      ticker1 ticker2 ticker3
date1  value1    None    None
date2    None  value2    None
date3    None    None  value3

Upvotes: 2

Related Questions