Python - 消除 numpy 数组或 pandas Z6A8064B53DF4794555570C53DF4794555570

Question

I have a pandas dataframe that currently looks like this

|Eriksson| NaN    | Boeser | NaN   |
| NaN    | McDavid| NaN    | NaN   |
| ...    | ...    | ...    | ...   |

I don't care whether its converted to a Numpy array or it remains a Data Frame, but I want an output object where the rows just consist of the non NaN values like this:

|Eriksson| Boeser|
|McDavid | NaN   |

( NaN because of the mismatched dimensions.) Is there any way to do this?

Answer 1

I think that this would do the trick for you:

df.apply(lambda x: pd.Series(x.dropna().values), axis=1)

Example:

>>> df = pd.DataFrame(np.random.randn(5,4))
>>> df.iloc[1,2] = np.NaN
>>> df.iloc[0,1] = np.NaN
>>> df.iloc[2,1] = np.NaN
>>> df.iloc[2,0] = np.NaN
>>> df
          0         1         2         3
0 -0.162388       NaN -0.299892  0.594846
1  3.165631 -1.190102       NaN -1.234934
2       NaN       NaN  0.885439 -1.714365
3 -1.622833 -1.319395 -1.716550 -0.517699
4  0.688479  0.576763  0.645344  0.708909
>>> df.apply(lambda x: pd.Series(x.dropna().values), axis=1)
          0         1         2         3
0 -0.162388 -0.299892  0.594846       NaN
1  3.165631 -1.190102 -1.234934       NaN
2  0.885439 -1.714365       NaN       NaN
3 -1.622833 -1.319395 -1.716550 -0.517699
4  0.688479  0.576763  0.645344  0.708909