如何将 `numpy.hstack()` 与 `numpy.ndarray` 数据类型一起使用？

Question

I have a numpy.DataFrame say df with 3 columns, col_1 , col_2 , col_3 . Data in col_1 are numpy.ndarray and looks like this: array([ 0.216, -0.290, 0.349])

How could I use np.hstack() to expand the DataFrame with columns consist of each data point in col_1 ?

ie Original DataFrame

                     col_1   col_2               col_3
------------------------------------------------------
0   [0.216, -0.290, 0.349]  NORMAL  N09_M07_F10_K001_1

Supposed DataFrame

                 col_3  col_2     0      1     2
------------------------------------------------
0   N09_M07_F10_K001_1 NORMAL 0.216 -0.290 0.349

I'd tried like this:

Supposed_DataFrame = pd.concat(
    [df[['label', 'filename']], 
     pd.DataFrame(np.hstack(df["signal"].values).T)
    ], 
    axis=1)

but the output was:

                 col_3  col_2     0
-----------------------------------
0   N09_M07_F10_K001_1 NORMAL 0.216

any simpler solutions will be appreciated

Answer 1

Check out this code:

import pandas as pd

dict_ = {
    'col_1': [[0.216, -0.290, 0.349]],
    'col_2': 'NORMAL',
    'col_3': 'N09_M07_F10_K001_1'
        }
df2 = pd.DataFrame(dict_)

supposed_DataFrame = pd.concat([df2[['col_3', 'col_2']], pd.DataFrame(df2['col_1'].to_list(), columns=[0,1,2])], axis=1)
print (supposed_DataFrame)

Method-2 : using basic steps:

row_1 = df2['col_1'][0]
for i in range(len(row_1)):
    df2[i] = row_1[i]

df2.drop('col_1', axis=1, inplace=True)

print(df2)