I have a dataFrame like above, with many more columns. I want to use "Animal" and "Sound" columns ( many more) to calculate 'score'.

Because there are many columns, using lambda does not seem possible.

Suppose I did

def assign_score(data):
   if data.Animal == "dog":
      data.score  = 1
   if data.Animal == "cat":
      data.score  = 2

   if data.Sound== "arf":
      data.score  = 10
   if data.Sound == "meow":
      data.score  = 20
df.Score = df.apply(assign_score)

I get a KeyError on this code. Is there a way to get it work?

Thanks

CodePudding user response：

import pandas as pd 
df=pd.DataFrame({'Animal':['dog','cat'], 'Sound': ['arf','meow'], 'score':[0,0]})
display(df)
# Give all of your column names in this list
colNames = ['Animal', 'Sound','score']

def assign_score(data):
    #print(type(data))
    if data[0] == "dog":
        score=data[2] 1
    if data[0] == "cat":
        score=2
    if data[1]== "arf":
        score=data[2] 10
    if data[1]== "meow":
        score=data[2] 20
    #print('score',score)
    return score
#df.Score = df.apply(assign_score)
df['score']=df.apply(lambda x: assign_score(x[colNames[0:]]), axis=1)
display(df)

CodePudding user response：

Try this

import pandas as pd

columns = ['animal', 'sound']

data = [('dog', 'arf'),
        ('cat', 'meow')]

df = pd.DataFrame(data, columns=columns)

score = {'dog' :  1, 'cat'  : 2,
         'arf' : 10, 'meow' :20}

df['score'] = df.apply(lambda x: sum([score[x[col]] for col in columns]), axis=1)

print(df)

  animal sound  score
0    dog   arf     11
1    cat  meow     22