I have a sample dataframe of sales:

For each state in my data frame, I would like to loop over each state and create a specific data frame:

state_lst = ['California', 'New York', 'Texas', 'Florida']

I know the long way would be to write out separate filtering step for each state: california_df = df[df['state'] == 'California']

But am looking for an efficient way to create a separate dataframe for each state:

for state in state_lst:
   state_df = df[df['state'] == state]
   state_df.groupby(['product_category'])[['total_revenue']].sum().reset_index()

My desired output is to create a specific dataframe for each state and then group that dataframe by sales in each product category.

Any suggestions?

CodePudding user response：

This will give you a list of dataframes containing only a unique state:

df_list = [df[df.state == unique_state] for unique_state in df.state.unique()]

CodePudding user response：

Maybe this:

df = df.groupby(['state', 'product_category'])[['total_revenue']].sum().reset_index()
[y for x, y in df.groupby('state', as_index=False)]