reco_implementation.py

# For the implementation, it was used Python decorator in Django REST framework to define a view function that can handle HTTP requests.

@api_view(['GET'])
def getRakingRecomm(request):
    category_arg = request.query_params.get('chosen_category')
    # the ML model returns the id of 3 ranked recommended items based on the category selected on the search bar
#    recommendation_dir = os.path.join(settings.BASE_DIR, 'm_l', 'recommendation')
#    sys.path.append(recommendation_dir)
#    from  recommendation_f import create_ranking_df
    id1, id2, id3 = create_ranking_df(category_arg)
    obj1 = Item.objects.get(id=id1)
    obj2 = Item.objects.get(id=id2)
    obj3 = Item.objects.get(id=id3)
    ranked_items = []
    serialized_obj1 = ItemSerializers(obj1).data
    ranked_items.append(serialized_obj1)
    serialized_obj2 = ItemSerializers(obj2).data
    ranked_items.append(serialized_obj2)
    serialized_obj3 = ItemSerializers(obj3).data
    ranked_items.append(serialized_obj3)

    return Response(ranked_items)