CODEWITHSUNDEEP
pythonpandas
anon_swe
anon_swe

Reputation: 9345

Pandas: Truth Value of a Series is Ambiguous

I have a Pandas DataFrame called df, containing a column called _type and one called avg_engaged_time. I want to look at the rows where _type is 0 and avg_engaged_time is between the 5th and 95th percentile. Here's my attempt so far:

First I filter based on _type:

original = result_df[result_df['_type'] == 0.0]

Then I find the percentiles:

low_original = original['_avg_engaged_time'].quantile(0.05)
high_original = original['_avg_engaged_time'].quantile(0.95)

Then I try to filter based on these percentiles:

original[original['_avg_engaged_time'] > low_original and original['_avg_engaged_time'] < high_original]

Unfortunately, I get this error:

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

I just want to use basic Boolean indexing to filter out rows that have an _avg_engaged_time less than the 5th percentile or greater than the 95th percentile...

Any ideas how to fix?

Thanks!

Upvotes: 1

Views: 627

Answers (1)

5agado
5agado

Reputation: 2454

You should use bitwise operator & instead of and. You are doing a logic operation between a list of boolean values, not on single ones.

so

original[(original['_avg_engaged_time'] > low_original) & (original['_avg_engaged_time'] < high_original)]

should work.

Upvotes: 3

Related Questions