For a list of users and activity dates, how to find the earliest date?

Question

I have a list of users and their activities, like this:

user	date	activity
Tom	1/1/21	Hop
Dick	1/2/21	Skip
Harry	2/2/21	Jump
Tom	1/3/21	Skip
Dick	1/4/21	Jump

I want to extract unique user names and the earliest activity date, to get a result like this:

user	first activity
Tom	1/1/21
Dick	1/2/21
Harry	2/2/21

I know I can create an array of unique usernames like this:

unique_users = user_actions[user].unique()

But I don't know how to turn that array of unique usernames into a dataframe with the first action date.

Answer 1

To get the desired result, you can do the following:

• Convert the date column to DateTime format for easy sorting and filtering
• Group by user and find the minimum date for each user

df['date'] = pd.to_datetime(df['date'])
result = (
    df.groupby("user")["date"]
    .agg(first_activity="min")
    .reset_index()
    .sort_values("first_activity")
)
print(result)

Output:

    user first_activity
2    Tom     2021-01-01
0   Dick     2021-01-02
1  Harry     2021-02-02