将 Pandas DataFrame 转换为没有 for 循环的一行

Question

I got a pandas dataframe like this:

    x   y   z
0   a   d   g
1   b   e   h
2   c   f   i

Now I want to convert it into a dataframe with a single row with each cell and row + column as column names:

    z_2 z_1 z_0 y_2 y_1 y_0 x_2 x_1 x_0
0   i   h   g   f   e   d   c   b   a

I know I can do it like this, but I need to runtime optimize the code, if possible without loops, etc.

df = pd.DataFrame({"x": ["a", "b", "c"],
                    "y": ["d", "e", "f"],
                    "z": ["g", "h", "i"]})
df.to_dict()
wantedRes = pd.DataFrame()
for key, value in df.items():
    for key2, value2 in value.items():
        wantedRes.insert(loc = 0, column = str(key) + "_" + str(key2),value = [value2] ) 

Answer 1

You can use .stack() for this:

s = df.stack()
df_new = pd.DataFrame([s.values],  columns=[f'{j}_{i}' for i, j in s.index])

Output:

  x_0 y_0 z_0 x_1 y_1 z_1 x_2 y_2 z_2
0   a   d   g   b   e   h   c   f   i

Answer 2

You can unstack , rework the index and convert to_frame :

s = df.unstack()
wantedRes = s.set_axis(s.index.map(lambda x: f'{x[0]}_{x[1]}'))[::-1].to_frame().T

output:

  z_2 z_1 z_0 y_2 y_1 y_0 x_2 x_1 x_0
0   i   h   g   f   e   d   c   b   a