CODEWITHSUNDEEP
pythonpandas
Osca
Osca

Reputation: 1694

Break a dataframe into several dataframes by using groupby and get_group

I'd like to break a dataframe into several dataframes base on the values in the Group column.

generate data

np.random.seed(0)

groups = ('Group 1', 'Group 2', 'Group 3')
means = ('Low', 'High', 'Mid')
index =  pd.MultiIndex.from_product([groups, means], names=['Group', 'Mean'])
values = np.random.randint(low=20, high=100, size=len(index))
data = pd.DataFrame(data={'val': values}, index=index).reset_index()

Group   Mean    val
Group 1 Low     64
Group 1 High    67
Group 1 Mid     84
Group 2 Low     87
Group 2 High    87
Group 2 Mid     29
Group 3 Low     41
Group 3 High    56
Group 3 Mid     90

I'm able to extract a single group

grouped = data.groupby(data.Group)
g1 = grouped.get_group("Group 1")

But I'd like to split the data into individual dataframes

Expected output:

data_Group1

Group   Mean    val
Group 1 Low     64
Group 1 High    67
Group 1 Mid     84

data_Group2

Group 2 Low     87
Group 2 High    87
Group 2 Mid     29

and same for Group 3

The real data is a lot bigger than this, so it will be great to automate the process. thanks

Upvotes: 1

Views: 143

Answers (1)

cs95
cs95

Reputation: 402413

You can call GroupBy and iterate over groups. Is a dictionary what you're looking for?

groups = {group_name : group for group_name, group in df.groupby('Group')}
groups['Group 1']

     Group  Mean  val
0  Group 1   Low   64
1  Group 1  High   67
2  Group 1   Mid   84

Upvotes: 1

Related Questions