Compare two columns in the same dataframe

Question

I want to compare two cloumns in the same dataframe and calculate the sum
this is my code:

sum=0
if df1['sma'] < df1['H+L'] :
        sum=sum + df1['H+L']
        print (sum)

this is my dataFrame:

df1 = pd.DataFrame({'sma':[1, 5, 2],
                    'Low':[1,2,3],
                    'H+L':[10, 20, 0]}, 
                    index=pd.date_range('2018-11-30', periods=3)).T
print (df1)
     2018-11-30  2018-12-01  2018-12-02
sma           1           5           2
Low           1           2           3
H+L          10          20           0

the output is a sum of all the rows. But always I get this error message

The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

I can't understand the issue.

Answer 1

I believe you need filter by boolean indexing with loc and use sum :

df1 = pd.DataFrame({'sma':[1, 5, 2],
                   'H+L':[10, 20, 0]})
print (df1)
   sma  H+L
0    1   10
1    5   20
2    2    0

print (df1.loc[df1['sma'] < df1['H+L'],'H+L']))
0    10
1    20
Name: H+L, dtype: float64

out = df1.loc[df1['sma'] < df1['H+L'],'H+L'].sum()
print (out)
30

If working with rows:

df1 = pd.DataFrame({'sma':[1, 5, 2],
                   'H+L':[10, 20, 0]}).T
print (df1)
      0   1  2
sma   1   5  2
H+L  10  20  0

out = df1.loc['H+L', df1.loc['sma'] < df1.loc['H+L']].sum()
print (out)
30

Answer 2

跟随应该起作用；

df1.apply(lambda x: x['H+L'] if x['sma']<x['H+L'] else 0, axis = 1).sum()