CODEWITHSUNDEEP
pythonpandas
csaladenes
csaladenes

Reputation: 1209

Reshape (stack) pandas dataframe based on predefined number of rows

I have a pandas dataframe which looks like one long row.

          0     1     2     3     4     5     6     7     8     9     10    11    12    13    14
________________________________________________________________________________________________
2010 |   0.1   0.5   0.5   0.7   0.5   0.5   0.5   0.5   0.9   0.5   0.5   0.8    0.3    0.3    0.6

I would like to reshape it as:

          0     1     2     3     4    
____________________________________
     |0| 0.1   0.5   0.5   0.7   0.5   
2010 |1| 0.5   0.5   0.5   0.9   0.5   
     |2| 0.5   0.8   0.3   0.3   0.6

I can certainly do it using a loop, but I'm guessing (un)stack and/or pivot might be able to do the trick, but I couldn't figure it out how...

Symmetry/filling up blanks - if the data is not integer divisible by the number of rows after unstack - is not important for now.

EDIT:

I coded up the loop solution meanwhile:

df=my_data_frame
dk=pd.DataFrame()
break_after=3
for i in range(len(df)/break_after):
    dl=pd.DataFrame(df[i*break_after:(i+1)*break_after]).T
    dl.columns=range(break_after)
    dk=pd.concat([dk,dl])

Upvotes: 0

Views: 59

Answers (1)

Liam Foley
Liam Foley

Reputation: 7822

If there is only one index (2010), this will work fine.

df1 = pd.DataFrame(np.reshape(df.values,(3,5)))
df1['Index'] = '2010'
df1.set_index('Index',append=True,inplace=True)
df1 = df1.reorder_levels(['Index', None])

Output:

           0    1    2    3    4
Index                           
2010  0  0.1  0.5  0.5  0.7  0.5
      1  0.5  0.5  0.5  0.9  0.5
      2  0.5  0.8  0.3  0.3  0.6

Upvotes: 1

Related Questions