CODEWITHSUNDEEP
pythonpandasdataframe
simon
simon

Reputation: 690

Tag values depending on multiple column conditions

Suppose that I have this minimal DataFrame to represent my needs:

    Tag   Year   Month   Snowed 
0    0    2001    Jan     True
1    0    2001    Feb     False 
2    0    2001    Mar     True
3    0    2002    Jan     True
4    0    2002    Feb     True
5    0    2002    Mar     False

How could I fill the Tag column such that, for each year, it is tagged as BAD if snowed in both Jan&Feb or GOOD otherwise?

Upvotes: 0

Views: 62

Answers (2)

BENY
BENY

Reputation: 323226

I would use transform, all, and ffill:

df['Tage']=df.loc[df.Month.isin(['Jan','Feb'])].groupby('Year')['Snowed'].transform('all').map({False:'Good',True:'Bad'})
df.ffill(inplace=True)
df
Out[262]: 
   Tag  Year Month  Snowed  Tage
0    0  2001   Jan    True  Good
1    0  2001   Feb   False  Good
2    0  2001   Mar    True  Good
3    0  2002   Jan    True   Bad
4    0  2002   Feb    True   Bad
5    0  2002   Mar   False   Bad

s=df.loc[df.Month.isin(['Jan','Feb'])].groupby('Year')['Snowed'].agg('all').map({False:'Good',True:'Bad'})

Then map back

df['Tage']=df.Year.map(s)

Upvotes: 1

Parmandeep Chaddha
Parmandeep Chaddha

Reputation: 484

Along with the above answer, this also works. This code might be less confusing though.

Since we don't care about the data in March as per your question: df = df[(df.Month != 'Mar')]

def True_Or_Not(df):
    Result = df['Snowed'].all(axis=None) == True
    df['Result'] = Result
    return df

Groupby Year and aggregate month column to determine whether all values are true

df  = df.groupby(['Year']).apply(lambda df: True_Or_Not(df)).reset_index().drop(columns = ['index'])
df['Tag'] = ['Good' if i== False else 'Bad' for i in df.Result.tolist()]

OUTPUT:

    Tag     Year    Month   Snowed  Result
0   Good    2001    Jan    True     False
1   Good    2001    Feb    False    False
2   Bad     2002    Jan    True     True
3   Bad     2002    Feb    True     True

Upvotes: 0

Related Questions