Pandas dataframe set column value based on nested if conditions

Question

I have 4 columns in the dataframe and would like to change all the values in column 2 based on the below condition:

if pd.isnull(df['COL2']) or df['COL2'] == "SOME_NAME":
    if pd.isnull(df['COL3']) == False:
        df['COL2'] = df['COL3']
    else:
        df['COL2'] = "DEFAULT"

so basically if col2 is empty or has some specific name replace it with col3 if col3 not empty else replace with default

I iterated over the df but the matching is not replacing all the values. I have some rows with col2 value still set to "SOME_NAME". Is there a simple way of doing this?

cs95 · Accepted Answer · 2018-02-05 17:53:41Z

1

Use assignment with loc, to only change those rows that meet your condition.

i = df['COL2'].isnull() | df['COL2'].eq('SOME_NAME')
j = df.loc[i, 'COL3']

df.loc[i, 'COL2'] = j.where(j.notnull(), 'DEFAULT')

edited Feb 5, 2018 at 17:53

answered Feb 5, 2018 at 17:24

cs95

406k106 gold badges745 silver badges798 bronze badges

Sign up to request clarification or add additional context in comments.

7 Comments

jezrael Over a year ago

Is not better chain third condition?

cs95 Over a year ago

@jezrael I don't know if they do the same thing, so I didn't chain it.

jezrael Over a year ago

OK, I try test it.

hmaxx Over a year ago

It does replace most of it but there are still rows with "SOME_NAME" value

cs95 Over a year ago

@hmaxx You might want to remove whitespaces, if any: df['COL2'].str.strip().eq('SOME_NAME')

|

jezrael · Accepted Answer · 2018-02-05 17:41:32Z

0

I think you need numpy.where with chanined conditions by | for or and & for and:

mask = (df['COL2'].isnull() | (df['COL2'].str.strip() == "SOME_NAME")) & df['COL3'].notnull()

df['COL2'] = np.where(mask, df['COL3'], "DEFAULT")

Comments

SteelFeather · Accepted Answer · 2018-02-05 18:02:31Z

0

The ternary operator was added in 2.5 i believe:

if (condition) else

sample dataframe:

df = pd.DataFrame({"COL2": ["first", None,"third","fourth", None], "COL3": ["first_","second_","third_","fourth_", None]})

so you can do something like this to replace value "third" or null :

df['COL2'] =df.apply(lambda row: row['COL3'] if ((not row['COL2'] or row['COL2'] == "third") and row['COL3']) else "DEFAULT", axis=1)

Output:

df
Out[21]: 
      COL2     COL3
0  DEFAULT   first_
1  second_  second_
2   third_   third_
3  DEFAULT  fourth_
4  DEFAULT     None

answered Feb 5, 2018 at 18:02

SteelFeather

1,76313 silver badges16 bronze badges

Collectives™ on Stack Overflow

Pandas dataframe set column value based on nested if conditions

3 Answers 3

7 Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

7 Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related