ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all() when comparing dataframe columns

Question

df['svc_port'] = np.where(
     min(df['sPort'],df['dPort']) <= 1024, 
     min(df['sPort'],df['dPort']), 
     df['dPort']
)

In the above code, min(df['sPort'],df['dPort']) <= 1024 - the same thing is given in Compare two columns using pandas. I am not using any logical operators. just checking a condition and replacing it's values.

Why am I still getting this error?

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

"I am not using any logical operators" - but you are, inside `min`. — user2357112, Mar 22 '18 at 03:17

score 2 · Accepted Answer · answered Mar 22 '18 at 03:19

You are looking for an element-wise min. The builtin min function only works for a single iterable, not multiple of them at a time.

What you're looking for is something along these lines, using np.minimum.

v = np.minimum(df['sPort'], df['dPort'])
df['svc_port'] = np.where(v <= 1024, v, df['dPort'])

ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all() when comparing dataframe columns

1 Answers1