calculating average depends of the number of variables of raw data in python

user_15136121 0
Viewed 51

I am stuck on calculating the average depends on the variables in python.

I have 8 variables as below.

Case 1.

Time         df1    df2   df3   df4   df5   df6  df7   df8
2020-01-01   220    250   235   215   221   221  220   253

In this case I can just calculate average like below

  df['dfaverage']=(df['df1']+df['df2']+df['df3']+df['df4']+df['df5']+df['df6']+df['df7']+df['df8'])/8

The output will be 229.4

but if one of the value is zero then how would i ignore that values and calculate? Case 2.

Time         df1    df2   df3   df4   df5   df6  df7   df8
2020-01-01   220    250   235   215   221   221  220    0

the output should be 226 for case2 but when I run the code I will get 197.8. How can I ignore 0 when calculating the average?

pandas
1 Answers

You can replace 0 to missing values and use mean:

df['dfaverage'] = df.replace(0, np.nan).mean(axis=1)

print (df)
       Time  df1  df2  df3  df4  df5  df6  df7  df8  dfaverage
0  2020-01-01  220  250  235  215  221  221  220    0      226.0

Or if need specify column names use list:

cols = [ 'df1', 'df2', 'df3', 'df4', 'df5', 'df6', 'df7', 'df8']
df['dfaverage'] = df[cols].replace(0, np.nan).mean(axis=1)

print (df)
         Time  df1  df2  df3  df4  df5  df6  df7  df8  dfaverage
0  2020-01-01  220  250  235  215  221  221  220    0      226.0
user_29010020
Related