INNER CODE UNIT · Python
make_user_sample
alteryx/open_source_demos · predict-next-purchase/process_data.py:7
def make_user_sample(orders, order_products, departments, products, user_ids, out_dir):
orders_sample = orders[orders["user_id"].isin(user_ids)]
orders_keep = orders_sample["order_id"].values
order_products_sample = order_products[order_products["order_id"].isin(orders_keep)]
try:
os.mkdir(out_dir)
except:
pass
order_products_sample.to_csv(os.path.join(out_dir, "order_products__prior.csv"), index=None)
orders_sample.to_csv(os.path.join(out_dir, "orders.csv"), index=None)
departments.to_csv(os.path.join(out_dir, "departments.csv"), index=None)
products.to_csv(os.path.join(out_dir, "products.csv"), index=None)
def main():
data_dir = "data"