我正在尋找轉置熊貓列并應用 Groupby
df = pd.DataFrame({'Code1' : ['X60', np.nan, 'X66', np.nan],
'Code2' : [np.nan, 'X64', 'X78', np.nan],
'Code3' : [np.nan, 'X66', 'X81', 'X59'],
'Code4' : [np.nan, np.nan, 'X38', 'X60']})
df
Code1 Code2 Code3 Code4
0 X60 NaN NaN NaN
1 NaN X64 X66 NaN
2 X66 X78 X81 X38
3 NaN NaN X59 X60
我怎樣才能達到這個預期的輸出？
Code NB
X38 1
X59 1
X60 2
X64 1
X66 2
X78 1
X81 1
uj5u.com熱心網友回復：
用于DataFrame.stackreshape 并洗掉缺失值和計數值，
Series.value_counts最后Series.sort_index使用2 列 DataFrame：Series.rename_axisSeries.reset_index
df = df.stack().value_counts().sort_index().rename_axis('Code').reset_index(name='NB')
print (df)
Code NB
0 X38 1
1 X59 1
2 X60 2
3 X64 1
4 X66 2
5 X78 1
6 X81 1
轉載請註明出處，本文鏈接：https://www.uj5u.com/yidong/490540.html