<p><img alt="" src="https://storage.yandexcloud.net/datasouls-ods/images/2020/04/17/3d8b48dd-4bfa-408a-b82d-3bedbe82d338/2-3.png" /></p>

<h2>Description</h2>

<p>You are required to build a fully-fledged recommendation system. Participants need to develop a high load web service that predicts the client&rsquo;s future purchase based on general information about the client and the client&rsquo;s purchase history. The web server responds in the form of a ranked list of <a href="https://en.wikipedia.org/wiki/Stock_keeping_unit">stock keeping units</a> (SKUs for short) that the client most likely is going to buy in the next purchase.</p>

<h2>Quality criteria</h2>

<p>Participants submit solutions in the form of an archived web service code. For this contest, the quality criterion is <strong>MNAP@30 (Mean Average Precision)</strong>: an average precision calculated for all requests and normalized afterward. The solution should conform to technical limitations (see page &ldquo;Data&rdquo;). Solutions that cannot handle the required load or respond slower than expected are also discarded as failed. During the test, all requests are sent in strict chronological order.</p>

<h2>Prizes</h2>

<p>Prize fund is 800&rsquo;000 RUR. The winning team will receive 300&rsquo;000 RUR; second place gets 150&rsquo;000 RUR; third place receives 100&rsquo;000 RUR; fourth and fifth places both get 75&rsquo;000 RUR. There is also a nomination for the best-open-sourced solution with a prize of 50&rsquo;000 RUR. Teams of up to 4 people from all over the world are allowed, regardless of actual physical location. You are entitled to participate in different teams in multiple Retail Hero contests.</p>

Build a recommender system service that will predict future purchases while withstanding the load

Private

https://storage.yandexcloud.net/ds-ods/files/media/meta_tag/file/a6a6037c2894/banner.jpg

description

X5 Retail Hero: Next Basket Prediction

Container

aprotopopov

ssh1

вжух-вжух и в продакшн

avolchek

vadimfb

anatoly

antitak

greenwolf

Пропустите наверх, пожалуйста

erqups

statz

Donchanin

chubasik

atikhonov

😎

mlenzovet

vkim

xchucx

man-in-turban

exotol

tixonmavrin

orcd

mihtw

SARRACENIA

maksimov.analytics

lisa_

Sergeif

latentvec

Крякнем, плюнем и надежно склеим скотчем

serjeeon

egorborisov

waniz

mylen_farmer

ekaterinae

timurdzhumakaev

JohnKeats

kg_tv

serzzh

shokh0

pkorobov

toohardforme

mkerimbay

imdxdd

alexb

Score

#	Teams (83)	Score	Last	Entries
1	aprotopopov	0.1483248823	5y	88
2	ssh1	0.1470261653	5y	114
3	вжух-вжух и в продакшн	0.145727632	5y	24
4	avolchek	0.1456443678	5y	124
5	vadimfb	0.1454425404	5y	119
6	anatoly	0.1440312277	5y	37
7	antitak	0.1437686753	5y	63
8	greenwolf	0.1434220496	5y	46
9	Рекомендации Красная Цена	0.1430190087	5y	71
10	Пропустите наверх, пожалуйста	0.1415798851	5y	16
11	8	0.1366046337	5y	45
12	erqups	0.1360933777	5y	26
13	statz	0.1355241008	5y	20
14	Donchanin	0.1355010433	5y	6
15	chubasik	0.134687587	5y	22
16	atikhonov	0.1342371231	5y	50
17	😎	0.133770888	5y	16
18	mlenzovet	0.133649891	5y	84
19	vkim	0.1333859235	5y	62
20	xchucx	0.1333786761	5y	14
21	man-in-turban	0.1328000604	5y	11
22	exotol	0.1327737425	5y	10
23	tixonmavrin	0.1327737425	5y	8
24	dk	0.1327737425	5y	9
25	orcd	0.1327737425	5y	2
26	mihtw	0.1327737425	5y	31
27	SARRACENIA	0.1327737425	5y	2
28	maksimov.analytics	0.1327194054	5y	8
29	lisa_	0.1318324513	5y	25
30	Sergeif	0.1312426863	5y	65
31	latentvec	0.1311689056	5y	12
32	Крякнем, плюнем и надежно склеим скотчем	0.1310684549	5y	16
33	serjeeon	0.1308989978	5y	15
34	egorborisov	0.1302589632	5y	7
35	waniz	0.1302589632	5y	3
36	if	0.1302589632	5y	50
37	mylen_farmer	0.1302589632	5y	2
38	ekaterinae	0.1302589632	5y	7
39	timurdzhumakaev	0.1302589632	5y	2
40	JohnKeats	0.1302589632	5y	15
41	kg_tv	0.1302589632	5y	3
42	serzzh	0.1302589632	5y	2
43	shokh0	0.1302589632	5y	3
44	pkorobov	0.1302589632	5y	2
45	toohardforme	0.1301959794	5y	2
46	mkerimbay	0.1301058801	5y	15
47	imdxdd	0.1295919968	5y	10
48	1	0.1291495246	5y	1
49	alexb	0.1291495246	5y	11
50	xph	0.1291495246	5y	5