How to fill in multiple boolean criteria in Pandas

Question

I found an alternate answer here link

My dataframe currently looks like this

          date  new_cases   new_deaths  new_tests   year    month   day weekday
0   2019-12-31  0.0                 0.0       NaN   2019    12      31   1
1   2020-01-01  0.0                 0.0       NaN   2020    1        1   2
2   2020-01-02  0.0                 0.0       NaN   2020    1        2   3

I want to pass an code which will average out the 'new_cases' for weekday and weekend my code currently look like this but I can only pass 1 condition which is ' == 6 '. I want to pass multiple criteria, for example, == (4,5,6)

covid_df[covid_df.weekday == 6].new_cases.mean()

Any clue?

Create a new column that codes whether a particular row is a weekday or not (e.g. is_weekday); then do a groupby over that column and then average. — navneethc, Jul 05 '21 at 11:03

score 1 · Accepted Answer · answered Jul 05 '21 at 11:08

1

I think you're looking for isin:

# you can use `loc` to access `new_cases` also
days = [4, 5, 6]
df.loc[df.weekday.isin(days), "new_cases"].mean()

which selects the rows whose weekday is in the days list; and then selects new_cases column.

answered Jul 05 '21 at 11:08

Mustafa Aydın

17,645
4
15
38

score 1 · Answer 2 · answered Jul 05 '21 at 11:17

You can groupby() date being a weekday or weekend. Following code sources and demonstrates this.

import requests
import io
import pandas as pd

dfall = pd.read_csv(io.StringIO(requests.get(
    "https://raw.githubusercontent.com/owid/covid-19-data/master/public/data/owid-covid-data.csv").text))
dfall["date"] = pd.to_datetime(dfall["date"])

dtlcols = ["date", "iso_code", "location", "continent"]
cols = ["new_cases", "new_deaths", "new_tests"]
dfall.loc[dfall["iso_code"].eq("GBR"), dtlcols + cols].groupby(
    dfall["date"].dt.dayofweek.isin([5, 6]), as_index=False
).agg({**{c: "last" for c in dtlcols}, **{c: "mean" for c in cols}})

	date	iso_code	location	continent	new_cases	new_deaths	new_tests
0	2021-07-02 00:00:00	GBR	United Kingdom	Europe	9536.18	288.555	456697
1	2021-07-04 00:00:00	GBR	United Kingdom	Europe	9214.97	204.614	393841

Thank you for your answer, but I am too beginner to understand that. Thanks anyway for your answer! This code could solve my question as well, but I'm not sure how it works, hope somebody will answer that for me. — FastBoi, Jul 05 '21 at 11:21
I'm a lazy typist :-). the only syntax sugar is building which columns we want in aggregate result with `agg()` — Rob Raymond, Jul 05 '21 at 11:22

How to fill in multiple boolean criteria in Pandas

2 Answers2