Pandas DataFrame 到列表列表字典

Question

I have CSV file with four columns: product_price, country_of_origin, product_quantity and brand_id. This is what csv looks like I want to create a dictionary where the key will be brand_id and the value will be a list of tuple/lists with other columns. Something like that:

some_dict = {
    1: 
    [(country_of_origin, product_quantity, product_price), 
     (country_of_origin, product_quantity, product_price),
     (country_of_origin, product_quantity, product_price)], 
    2: 
    [(country_of_origin, product_quantity, product_price), 
     (country_of_origin, product_quantity, product_price)],
    3:
    [(country_of_origin, product_quantity, product_price), 
     (country_of_origin, product_quantity, product_price)]
}

Is it possible to create such structure with pandas? I've tried using {x[3]: x[0:] for x in df.itertuples(index=False)} but it returns only one value per brand_id:

{1: (200, 'Kenya', 19), 3: (40, 'South Africa', 40), 2: (350, 'Turkey', 64)}

Answer 1

You can use dict comprehension with groupby brand_id and DataFrame.iterrows :

some_dict = {k: [(co, pq, pp) for _, (pp, co, pq, _) in x.iterrows()]
             for k, x in df.groupby('brand_id')}

[out]

{1: [('Kenya', 19, 200), ('Turkey', 25, 35), ('Jordan', 53, 16)],
 2: [('Turkey', 64, 350), ('Jordan', 24, 80)],
 3: [('South Africa', 5, 40), ('Oman', 8, 63)]}