stat.md

# pandas dataframe之apply

apply的使用


## template

```python
import pandas as pd
import numpy as np

def add_val(num):
    if num > 0:
        return 1
    else:
        return 0


if __name__ == '__main__':
    data = {
        'cloumn_one': pd.Series(np.random.randint(-10, 10, size=5)),
        'cloumn_two': pd.Series(np.random.randint(0, 10, size=5)),
        'cloumn_three': pd.Series(np.random.randint(0, 10, size=5))
    }
    data_df = pd.DataFrame(data)

    data_df['label_one'] = data['cloumn_one'].apply(add_val)

    print(data_df)
```

## 答案

```python
import pandas as pd
import numpy as np

if __name__ == '__main__':
    data = {
        'cloumn_one': pd.Series(np.random.randint(-10, 10, size=5)),
        'cloumn_two': pd.Series(np.random.randint(0, 10, size=5)),
        'cloumn_three': pd.Series(np.random.randint(0, 10, size=5))
    }
    data_df = pd.DataFrame(data)

    data_df['label_one'] = data['cloumn_one'].apply['num'>1]

    print(data_df)
```

## 选项

### A

```python
import pandas as pd
import numpy as np

def add_val(num):
    if num > 0:
        return 1
    else:
        return 0


if __name__ == '__main__':
    data = {
        'cloumn_one': pd.Series(np.random.randint(-10, 10, size=5)),
        'cloumn_two': pd.Series(np.random.randint(0, 10, size=5)),
        'cloumn_three': pd.Series(np.random.randint(0, 10, size=5))
    }
    data_df = pd.DataFrame(data)

    data_df['label_one'] = data['cloumn_one'].apply(add_val)

    print(data_df)
```

### B

```python
import pandas as pd
import numpy as np

if __name__ == '__main__':
    data = {
        'cloumn_one': pd.Series(np.random.randint(-10, 10, size=5)),
        'cloumn_two': pd.Series(np.random.randint(0, 10, size=5)),
        'cloumn_three': pd.Series(np.random.randint(0, 10, size=5))
    }
    data_df = pd.DataFrame(data)

    data_df['label_one'] = data['cloumn_one'].apply(lambda num: 1 if num>0 else 0)

    print(data_df)
```

### C

```python
import pandas as pd
import numpy as np

if __name__ == '__main__':
    data = {
        'cloumn_one': pd.Series(np.random.randint(-10, 10, size=5)),
        'cloumn_two': pd.Series(np.random.randint(0, 10, size=5)),
        'cloumn_three': pd.Series(np.random.randint(0, 10, size=5))
    }
    data_df = pd.DataFrame(data)

    def add_val(num):
        return 1 if num>0 else 0

    data_df['label_one'] = data['cloumn_one'].apply(add_val)

    print(data_df)
```