Write 'Big', 'Medium', 'Small' to new column based on what the biggest size in the row is

Question

Here is my example data frame:

|   Cat     |   Dog     |  Bird     |   Cow     |   Pig     |  
|:------:   |:------:   |:-----:    |:------:   |:------:   |  
|   Big     |   Big     | Small     | Medium    | Medium    |  
|  Small    |  Small    | Small     |  Small    |  Small    |  
| Medium    | Medium    | Small     | Medium    |  Small    |

I would like to create another column 'Size' with the value 'Big' if any of the animals are Big or else 'Medium' if any are Medium or else 'Small' if all the animals are only Small. The result Im looking for is seen below.

|   Cat     |   Dog     |  Bird     |   Cow     |   Pig     | Size      |  
|:------:   |:------:   |:-----:    |:------:   |:------:   |--------   |  
|   Big     |   Big     | Small     | Medium    | Medium    | Big       |  
|  Small    |  Small    | Small     |  Small    |  Small    | Small     |  
| Medium    | Medium    | Small     | Medium    |  Small    | Medium    |

I've tried

df['Size'] = df[['Cat', 'Dog', 'Bird', 'Cow','Pig' ].map(lambda x: 'Big' if 'Big' in x
                                                        else 'Medium' if 'Medium' in x 
                                                        else 'Small' if 'Small' in x)

I've also tried this and just added done it for all animals but it obviously only returns the results for the last one.

df['Size'] = df['Cat'].map(lambda x: 'Big' if 'Big' in x
                                                        else 'Medium' if 'Medium' in x
                                                        else 'Small' if 'Small' in x)

Please see [this question](https://stackoverflow.com/questions/20109391/how-to-make-good-reproducible-pandas-examples) on how to ask questions involving pandas dataframes. — vlizana, Jun 26 '20 at 03:59

score 0 · Accepted Answer · answered Jun 26 '20 at 03:15

0

You could try this:

df['Size']=['Big' if 'Big' in row else('Medium' if 'Medium' in row else('Small' if 'Small' in row else ''))for row in df.to_records()]

answered Jun 26 '20 at 03:15

MrNobody33

6,413
7
19

Thank you, this worked and was relatively simple :) – big_soapy Jun 26 '20 at 21:01

score 0 · Answer 2 · answered Jun 26 '20 at 05:46

Create a series of checks according to your conditions, then apply numpy select to create your new column

cond1 = df.eq("Big").any(axis=1)
cond2 = df.eq("Medium").any(axis=1)
cond3 = df.eq("Small").all(axis=1)
condlist = [cond1, cond2, cond3]
choicelist = ["Big", "Medium", "Small"]
df["Size"] = np.select(condlist, choicelist)

Write 'Big', 'Medium', 'Small' to new column based on what the biggest size in the row is

2 Answers2