CODEWITHSUNDEEP
pythonpandasdataframe
dazai
dazai

Reputation: 716

pandas.core.base.DataError: No numeric types to aggregate when trying to get the mean with groupby

I'm getting the error: "pandas.core.base.DataError: No numeric types to aggregate" when I try to get the mean of a column values based on the values of another column.

This is how the columns look:

   PRESSEDKEY  PALABRA   COLOR KEYCORR   RT CORRECT
90           v    ABUSO     red       r  496       N
108          v  FRACASO    blue       a  168       N
138          v   MORGUE   green       v  106       Y
150          v    ENOJO     red       r    0       N
156          v     ODIO   green       v   25       Y

I'm trying to get the mean RT of the values "N" and "Y" in the CORRECT column.

This is what I'm trying and gives me the mentioned error:

mean_emo = df_emowords.groupby('CORRECT')['RT'].mean()

How can I solve this?

Upvotes: 0

Views: 187

Answers (1)

Celius Stingher
Celius Stingher

Reputation: 18367

DataError: No numeric types to aggregate

Can be replicated when the column / data passed to the groupby object is not of numeric type. You have to previously convert it to numeric:

df_emowords['RT'] = pd.to_numeric(df_emowords['RT'],errors='coerce)

And now it can be correctly ran:

mean_emo = df_emowords.groupby('CORRECT')['RT'].mean()

Upvotes: 1

Related Questions