Append constant string to first row a dataframe in Pandas

Question

The objective is to append a string at the first row of a df.

The snippet below answer the objective

import pandas as pd
df = pd.DataFrame({'A': [11, 21, 31],
               'B': [12, 22, 32],
               'C': [13, 23, 33]},
              index=['ONE', 'TWO', 'THREE'])

df=df.astype ( str )
all_columns = list ( df )

df_new=pd.DataFrame(['test']*len(all_columns)).transpose()
df_new.columns=all_columns
df_new = df_new.astype ( str )
df33= pd.concat ( [df_new, df], axis=0 )

But I wonder whether there is more compact way that the proposed solution.

Answer 1

append and concat are concise, but it's much more efficient to prepend the row as a list insertion:

data = df.to_numpy().tolist()
data.insert(0, ['test'] * len(df.columns))
pd.DataFrame(data, index=[0] + df.index.tolist(), columns=df.columns)

#           A     B     C
# 0      test  test  test
# ONE      11    12    13
# TWO      21    22    23
# THREE    31    32    33

Timings:

%%timeit
data = df.to_numpy().tolist()
data.insert(0, ['test'] * len(df.columns))
pd.DataFrame(data, index=[0] + df.index.tolist(), columns=df.columns)

# 413 µs ± 57 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each)

%%timeit
pd.DataFrame({c:['test'] for c in df.columns}).append(df)

# 807 µs ± 83.8 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each)

Answer 2

Here is a solution you can give it a try,

pd.DataFrame({c:['test'] for c in df.columns}).append(df)

---

          A     B     C
0      test  test  test
ONE      11    12    13
TWO      21    22    23
THREE    31    32    33

Answer 3

Just a question, why not using MultiIndex to keep your column with the right dtype?

df.columns = pd.MultiIndex.from_product([df.columns, ['test']])

>>> df
         A    B    C
      test test test
ONE     11   12   13
TWO     21   22   23
THREE   31   32   33