Count number of symptoms from dataset

user_10533966 0
Viewed 55

For the dataset I am working with I am trying to create a new column called NumberSymptoms which is the number of symptoms someone has. To do this I am trying to go through the columns in each row and if it is a yes for a specific one, add that to the count, and then eventually it will come to a total number.

So it should eventually be something like

Cough Myalgia Headache SoreThroat Fatigue NumberSymptoms
Yes Yes No Yes No 3
No Yes Yes Yes Yes 4

And so on for the rest of the rows.

I have tried to make a function for this:

number = 0
def count_symptoms(Cough, Myalgia, Headache, SoreThroat, Fatigue):
    if Cough == "Yes":
        number +=1
    elif Myalgia == "Yes":
        number +=1 
    elif Headache == "Yes":
        number +=1 
    elif SoreThroat == "Yes":
        number +=1
    elif Fatigue == "Yes":
        number +=1
    return number
    
df["NumberSymptoms"] = count_symptoms(df["Cough"], df["Myalgia"], df["Headache"], df["SoreThroat"], df["Fatigue"])

But I am getting the error:

ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all(). And I'm not sure why and I'm sure there must be a better way to do this, I'm just not sure what it is

pandas python
5 Answers

Try this:

df['NumberSymptoms'] = (df[['Cough','Myalgia','Headache','SoreThroat','Fatigue']] == 'Yes').sum(axis=1)
user_136833600

If you want to check one value, you can use

df['res'] = df[cols].eq('Yes').sum(axis=1)

If you want to check multiple values, you can try applymap and check if value if list.

cols = ['Cough', 'Myalgia', 'Headache', 'SoreThroat', 'Fatigue']
vals = ['Yes', 'No']

df['res'] = df[cols].applymap(lambda x: x in vals).sum(axis=1)
user_103151630

Check if this fits your problem.

df["NumberSymptoms"] = df.apply(lambda row: sum(row=='Yes') ,axis=1)
user_110054370

I went about it a different way and managed to get what I needed:

number = 0
symptomList = [df.Cough, df.Myalgia, df.Headache, df.SoreThroat, df.Fatigue]
for symptom in symptomList:
    number+= symptom.str.count("Yes")
    number+= symptom.str.count("Dry")
    number+= symptom.str.count("Other")


df["NumberSymptoms"] = number

I hadn't read the specification properly before and for 1 of the columns I needed to look for either Dry or Other.

But getting the columns into a list and then going through each one and looking for the words seems to work for me

user_105339660

An elegant solution would be to map the values then sum:

cols = ['Cough', 'Myalgia', 'Headache', 'SoreThroat', 'Fatigue']
df[[cols]].map({'Yes': True, 'No': False}).sum(axis=1)
user_105219590
Related