Dynamically create DF and filter with loop

Question

What is the best practice to loop this?

I want to filter and dynamically create new DFs

df_aus19 = df_mat.loc[(df_temp['Year'] == 2020 ) & (df_mat['Country'] == 'AUS')]
df_aus20 = df_mat.loc[(df_temp['Year'] == 2020 ) & (df_mat['Country'] == 'AUS')]
df_aus21 = df_mat.loc[(df_temp['Year'] == 2021 ) & (df_mat['Country'] == 'AUS')]
df_aus22 = df_mat.loc[(df_temp['Year'] == 2022 ) & (df_mat['Country'] == 'AUS')]
df_aus23 = df_mat.loc[(df_temp['Year'] == 2023 ) & (df_mat['Country'] == 'AUS')]

AS per code

In my answer I am assuming the first filter should be 2019 as implied by the naming of the corresponding variable. — Scriddie
– Scriddie, Commented Jun 4, 2019 at 13:12

Scriddie · Accepted Answer · 2019-06-04 13:22:35Z

1

A more pythonic way to solve your problem than by loops and dynamically creating global variables would be a list comprehension:

years = list(range(2019, 2024))
df_aus_list = [df_mat.loc[(df_temp['Year'] == i) & (df_mat['Country'] == 'AUS')] for i in years]

If you really need to produce the same result as your code you could dynamically create new global variables like so:

years = list(range(2019, 2024))
for i in years:
    globals()["df_aus"+str(i % 100)] = df_mat.loc[(df_temp['Year'] == i ) & (df_mat['Country'] == 'AUS')]

edited Jun 4, 2019 at 13:22

answered Jun 4, 2019 at 13:05

Scriddie

3,9491 gold badge14 silver badges20 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Serge Ballesta · Accepted Answer · 2019-06-04 13:41:12Z

1

Don't. Just don't. When you find yourself ready to dynamically create new variables, just refrain: it is hard to obtain and harder to maintain. Just use containers like dictionaries or lists.

And the common way to get that in pandas is group_by:

dfs = dict()
for (year, country), sub in df.groupby(['Year', 'Country']):
  # if 2020 <= year <= 2023 and county == 'AUS':     # filter optionaly
      dfs[(year, country)] = sub

You can then get the dataframe for AUS and 2021 as dfs[(2021, 'AUS')]

answered Jun 4, 2019 at 13:41

Serge Ballesta

150k13 gold badges137 silver badges267 bronze badges

1 Comment

Pat Tierney Over a year ago

Nice trick with dfs..so each Value corresponds to dataframe referenced by the Key (2019, AUS')...etc.

Collectives™ on Stack Overflow

Dynamically create DF and filter with loop

2 Answers 2

Comments

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related