转换数据框以列出和删除 NaN 值

transform a dataframe to list ans delete NaN values

嗨,我有以下数据集

df = pd.DataFrame({'fruits': ['orange', 'mango', 'apple', 'grapes', 'NaN', 'mango'],
                   'price': ['40', '80', 'NaN', '40', '30', '80']
                   })

    fruits  price
0   orange  40
1   mango   80
2   apple   NaN
3   grapes  40
4   NaN     30

我想要 return 一个没有 NaN 值的列表。所以我使用以下代码:

dfd= [[y for y in x if pd.notna(y)] for x in df.values.tolist()]

但是,NaN 值仍然存在

[['orange', '40'],
 ['mango', '80'],
 ['apple', 'NaN'],
 ['grapes', '40'],
 ['NaN', '30'],
 ['mango', '80']]

有什么想法吗?

因为NaN是字符串pd.notnareturnFalse,如果需要删除'NaN's字符串使用:

dfd= [[y for y in x if Y != 'NaN'] for x in df.values.tolist()]

如果将 NaNs 个字符串转换为缺失值:

df = df.replace('NaN', np.nan)
dfd= [[y for y in x if pd.notna(y)] for x in df.values.tolist()]