The raw dataset consists of 9 different columns

date - is the timestamp when the purchase was done

id_doc - receipt's id

id_order - is the order's id. Order's id is equal to 0 when the purchase was done offline, otherwise online.

id_card - is the unique id of customer.

id_tov - is the unique id of product's id

id_kontr - distributor's id

quantity - the amount of the same product in receipt which was purchased

sum - price of the product

is_green - shows that product is under discount or not