使用必须在另一列的特定行上计算的公式填充 Pandas 数据框行，其值为 0 或 NaN

Question

我有一个日期框，其中 "price" 列中的值根据 "quantity" 和 "year" 列中的值而不同.例如，对于等于 2 的数量，我在 2017 年的价格等于 2，在 2018 年的价格等于 4。我想用 2018 年的值填充 2019 年的行，这些行具有 0 和 NaN 值。

df = pd.DataFrame({
    'quantity': pd.Series([1,2,3,4,5,6,7,8,9,1,2,3,4,5,6,7,8,9,1,2,3,4,5,6,7,8,9]),
    'year': pd.Series([2017,2017,2017,2017,2017,2017,2017,2017,2017,2018,2018,2018,2018,2018,2018,2018,2018,2018,2019,2019,2019,2019,2019,2019,2019,2019,2019,]),
    'price': pd.Series([1,2,3,4,5,6,7,8,9,2,4,6,8,10,12,14,16,18,np.NaN,np.NaN,0,0,np.NaN,0,np.NaN,0,np.NaN])
})

如果我不使用 2018 年的值，而是计算 2017 年和 2018 年之间的 平均值 呢？

我尝试重新适应将其应用于第一个案例（以应用 2018 年的数据），但它不起作用：

df['price'][df['year']==2019].fillna(df['price'][df['year'] == 2018], inplace = True)

你能帮帮我吗？

预期的输出应该是如下所示的数据帧：

Df 具有 2018 年的值

df = pd.DataFrame({
    'quantity': pd.Series([1,2,3,4,5,6,7,8,9,1,2,3,4,5,6,7,8,9,1,2,3,4,5,6,7,8,9]),
    'year': pd.Series([2017,2017,2017,2017,2017,2017,2017,2017,2017,2018,2018,2018,2018,2018,2018,2018,2018,2018,2019,2019,2019,2019,2019,2019,2019,2019,2019,]),
    'price': pd.Series([1,2,3,4,5,6,7,8,9,2,4,6,8,10,12,14,16,18,2,4,6,8,10,12,14,16,18])
})

Df 的值为 2017 年和 2018 年之间的平均值

df = pd.DataFrame({
    'quantity': pd.Series([1,2,3,4,5,6,7,8,9,1,2,3,4,5,6,7,8,9,1,2,3,4,5,6,7,8,9]),
    'year': pd.Series([2017,2017,2017,2017,2017,2017,2017,2017,2017,2018,2018,2018,2018,2018,2018,2018,2018,2018,2019,2019,2019,2019,2019,2019,2019,2019,2019,]),
    'price': pd.Series([1,2,3,4,5,6,7,8,9,2,4,6,8,10,12,14,16,18,1.5,3,4.5,6,7.5,9,10.5,12,13.5])
})

Answer 1

这是一种用 2017 和 2018 的平均值填充的方法。首先按数量对前一年的数据进行分组，然后汇总平均值：

m = df[df.year.isin([2017, 2018])].groupby('quantity').price.mean()

使用 set_index to set the quantity column as index, replace 0s by NaNs and use fillna 也接受字典来根据索引映射值：

ix = df[df.year.eq(2019)].index
df.loc[ix, 'price'] = (df.loc[ix].set_index('quantity').price
                        .replace(0, np.nan).fillna(m).values)

    quantity  year  price
0          1  2017    1.0
1          2  2017    2.0
2          3  2017    3.0
3          4  2017    4.0
4          5  2017    5.0
5          6  2017    6.0
6          7  2017    7.0
7          8  2017    8.0
8          9  2017    9.0
9          1  2018    2.0
10         2  2018    4.0
11         3  2018    6.0
12         4  2018    8.0
13         5  2018   10.0
14         6  2018   12.0
15         7  2018   14.0
16         8  2018   16.0
17         9  2018   18.0
18         1  2019    1.5
19         2  2019    3.0
20         3  2019    4.5
21         4  2019    6.0
22         5  2019    7.5
23         6  2019    9.0
24         7  2019   10.5
25         8  2019   12.0
26         9  2019   13.5

使用必须在另一列的特定行上计算的公式填充 Pandas 数据框行，其值为 0 或 NaN

Fill Pandas dataframe rows, whose value is a 0 or NaN, with a formula that have to be calculated on specific rows of another column

python

pandas

function

fillna