问题

我需要在包含多个个体的数据集中添加一个次要ID，以确保条目唯一。为此，我尝试使用`np.where()`，但在实施后我意识到每次都在覆盖上一个条目。这是原始方法的示例：

```python
df = pd.DataFrame({'Example':['1','2','3','4']})

df['Add'] = ''

df['Add'] = np.where(df['Example']== '1', 'One','')
df['Add'] = np.where(df['Example']== '2', 'Two','')
df['Add'] = np.where(df['Example']== '3', 'Three','')
df['Add'] = np.where(df['Example']== '4', 'Four','')

df.head()

作为解决办法，我尝试添加了str.contains('')，认为当字符串为空时会评估为True，并且只在这种情况下插入新字符串。如下：

df = pd.DataFrame({'Example':['1','2','3','4']})

df['Add'] = ''

df['Add'] = np.where(df['Example'].str.contains('')== '1', 'One','')
df['Add'] = np.where(df['Example'].str.contains('')== '2', 'Two','')
df['Add'] = np.where(df['Example'].str.contains('')== '3', 'Three','')
df['Add'] = np.where(df['Example'].str.contains('')== '4', 'Four','')

df.head()

在这种情况下，所有内容都被填充为空字符串...

有没有一种简单的方法在使用np.where()写入之前检查单元格是否为空？

<details>
<summary>英文:</summary>

I need to add a secondary ID in a dataset with unique entries for multiple individuals. To do so I am trying to use `np.where()`, after I implemented I realized I am overwriting the last entry each time. This an example of the original approach:

df = pd.DataFrame({'Example':['1','2','3','4']})

df['Add'] = ''

df['Add'] = np.where(df['Example']== '1', 'One','')
df['Add'] = np.where(df['Example']== '2', 'Two','')
df['Add'] = np.where(df['Example']== '3', 'Three','')
df['Add'] = np.where(df['Example']== '4', 'Four','')

df.head()

As a work around I tried adding `str.contains('')` thinking that would evaluate `True` when string is empty and only insert new string in that case. As below:

df = pd.DataFrame({'Example':['1','2','3','4']})

df['Add'] = ''

df['Add'] = np.where(df['Example'].str.contains('')== '1', 'One','')
df['Add'] = np.where(df['Example'].str.contains('')== '2', 'Two','')
df['Add'] = np.where(df['Example'].str.contains('')== '3', 'Three','')
df['Add'] = np.where(df['Example'].str.contains('')== '4', 'Four','')

df.head()

In that instance everything is being filled with an empty string... 

Is there a simple method to check if a cell is empty before writing with `np.where()`? 



</details>


# 答案1
**得分**: 2

使用 [`map`][1] 函数：

```python
dmap = {'1': 'One', '2': 'Two', '3': 'Three', '4': 'Four'}

df['Add'] = df['Example'].map(dmap).fillna('')

输出：

>>> df
  Example    Add
0       1    One
1       2    Two
2       3  Three
3       4   Four

dmap = {'1': 'One', '2': 'Two', '3': 'Three', '4': 'Four'}

df['Add']  = df['Example'].map(dmap).fillna('')

>>> df
  Example    Add
0       1    One
1       2    Two
2       3  Three
3       4   Four