How to create a dictionary of dataframes for each column category

Question

I have a dataframe that looks like this:

df = pd.DataFrame([[201801, 0.5, 273.4, 'Fleet'], [201801, 0.34, 277.4, 'Drake'], [201801, 0.75, 255, 'Bay'], [201802, 0.97, 244.4, 'Fleet'], [201802, 0.54, 267.4, 'Drake'], [201802, 0.89, 235, 'Bay']], columns = ['time', 'windspeed', 'winddir', 'site_name'])
df

I want to create a dictionary of dataframes where the dictionary key is the site_name column and then the value is rest of the dataframe (i.e. the other 3 columns).

How can I do this please?

I do create this dataframe from combining 4 arrays into a dataframe earlier on, so if it is easier, then I could create the dictionary of dataframes from the arrays instead?

Asked By: splodyn

||

Source

Answer 1

GroupBy object, when iterated over, gives back the grouper keys along with the group dataframe; therefore a dictionary comprehension is possible. The grouper is also included in that subframe, so we drop that:

grouper = "site_name"

d = {name: sub_df.drop(columns=grouper) for name, sub_df in df.groupby(grouper)}

to get

>>> d

{"Bay":      time  windspeed  winddir
 2  201801       0.75    255.0
 5  201802       0.89    235.0,

 "Drake":      time  windspeed  winddir
 1  201801       0.34    277.4
 4  201802       0.54    267.4,

 "Fleet":      time  windspeed  winddir
 0  201801       0.50    273.4
 3  201802       0.97    244.4}

Answered By: Mustafa Aydın

How to create a dictionary of dataframes for each column category

Question:

Answers: