<img alt="" src="https://storage.yandexcloud.net/datasouls-competitions/sberbank-covid19-forecast/covid-scr-2.png" />

The participants need to build an algorithm for the most accurate prediction of the number of reported cases of COVID-19 dynamics for the next seven days. 
The competition aims at drawing attention to the pandemic forecasts building. It may probably help to find problems in the data sources or build a useful forecast using the most reliable data.

 
<a href="http://opendatascience.slack.com/archives/C011745MET0">The competition channel in&nbsp;Slack ODS:&nbsp;#proj_covid</a> 
<a href="http://covidforecast.sberbank.ai/">Description of the competition in Russian</a>

<img alt="" src="https://storage.yandexcloud.net/datasouls-competitions/sberbank-covid19-forecast/covid-scr-5.png" />

The main goal of the contest is open science development support, forecasting techniques development, and sharing experience in building predictive models. The tasks like that organized as public benchmarking allow scientists, researchers, and engineers to test and compare different approaches and together develop the best practices, at the same time making them available to the entire research community.

The next goal of the contest is to draw attention to the problem of pandemic surveillance and qualitative modeling of its spreading worldwide. The development of open qualitative predictive models that allow assessment of the immediate dynamics of epidemics supports future quick decision-making and effective responses aimed to prevent the next pandemic spread. 
 
<a href="https://storage.yandexcloud.net/datasouls-competitions/sberbank-covid19-forecast/COVID19_Data_Challenge_Rules_RU%20-_17.pdf">Contest rules RU</a> 
<a href="https://storage.yandexcloud.net/datasouls-competitions/sberbank-covid19-forecast/COVID-19_Rules_EN_17.pdf">Contest rules EN</a> 
<a href="https://www.sberbank.ru/ru/pol_pd">Processing of Personal Data Sberbank RU</a>

<h3>Prizes</h3>

<img alt="" src="https://storage.yandexcloud.net/datasouls-competitions/sberbank-covid19-forecast/covid-scr-1.png" />

The winners are determined in three stages:

&nbsp;

<ul>
	<li>Stage 1: for the week 13.04 - 19.04, the winners will be determined on Tuesday 21.04 (solutions sending deadline is 12.04, 11:59&nbsp;p.m., GMT +3 (Moscow)). 
	&nbsp;</li>
	<li>Stage 2: for the week 20.04 - 26.04, the winners are determined on Tuesday 28.04 (solutions sending deadline is 19.04, 11:59&nbsp;p.m., GMT +3 (Moscow)). 
	In the second stage of the competition, the results will be announced both according to country forecasts and for Russian regions. 
	Second stage prizes will be given to those who make the best forecasts for Russian regions. 
	&nbsp;</li>
	<li>Stage 3: for the week 27.04 - 03.05 winners are determined on Tuesday 05.05 (solutions sending deadline is 26.04, 11:59&nbsp;p.m., GMT +3 (Moscow)). 
	In the third (final) stage of the competition, the results will be announced both according to country forecasts and for Russian regions. 
	This stage prizes will be given to those who make the best forecasts for Russian regions. 
	&nbsp;</li>
</ul>

<h3>FAQ</h3>

What&nbsp;period is forecasted?

The forecast is produced for each day until the end of the competition. The intermediate results are given for one week. It may seem like a too-short planning horizon for such a long-distance problem. However, the epidemic&rsquo;s dynamic is strongly influenced by the quarantine measures taken in each country. It is not possible to predict the exact dates of their implementation as well as their effectiveness.&nbsp; 
Moreover, as of the contest starting date, there is no complete data on all the measures implemented, and the impact of the implemented measures has yet to be assessed.

Why predict if there is no confidence in the data validity?

Considering the contest task limitations, the algorithms and solutions developed in its course can still be useful, especially given the increase in data reliability. The competition is, first of all, a public benchmark of predictive models for COVID-19 epidemic data. The participants should focus primarily on modeling rather than on engineering aspects and limitations of systems and solutions, just like their models had to be immediately used in the real environment.

How were the evaluation metrics chosen? 
 
Task features cause the choice of metrics. First, the number of infected people grows exponentially. Second, the general development of the epidemic dynamics in any of the countries is more important than focusing on predicting the exact number of infected in most-pandemic countries. And third, different countries have different standards and approaches to COVID-19 testing, as well as different stages of the epidemic, so the problem needs a sustainable path.

Please note that the metrics may change during the contest. The final version will be announced no later than a week before the end of the competition. 
 
The organizers are waiting for suggestions for improvement from the Open Data Science community.

Use any data you can find to predict the future increase in the number of reported cases of COVID-19

Everyday best results

1 stage

2 stage

3 stage

https://storage.yandexcloud.net/datasouls-ods/static/meta_tags/ce9f2626-c9fe-49a5-b5ae-3f63e59e060e_badge-covid_DLE8PQ9.jpg

Use any data you can find to predict the future increase of the number of reported cases of COVID-19.

#	Teams (111)	Точность
1	Niw	0.3726203475
2	ZFTurbo (Роман Соловьев)	0.3963948074
3	solemn_leader	0.3995833591
4	Школьники Нонны	0.4110715196
5	Philipp Lozovoy	0.4185969953
6	Siberian GANs	0.4304787252
7	Квентин Карантино	0.4374997843
8	Topspin26	0.4531676923
9	Alexei Chernobrovov	0.455656682
10	Andrew Shevelev	0.4642920838
11	KJ	0.4703927764
12	Роман Пьянков	0.4941429448
13	Олег Шаповалов	0.4996154069
14	Renat Ramazanov	0.5143265391
15	Кирилл Рыбкин	0.5268481508
16	how	0.5333988307
17	ihvh@github	0.5431243698
18	bat_soup	0.555528638
19	Medical Advisers Group	0.5600744662
20	Jupyter Invaders	0.5613930541
21	a18	0.5670564946
22	Последний Боярин	0.5675621738
23	drts	0.5721771588
24	Alex	0.5762040914
25	Ivan Komarov	0.5867689226
26	Babusya	0.5931349983
27	Victor Zaguskin	0.5946543435
28	Михаил	0.6017631196
29	Vladislav	0.6179009738
30	JK	0.621123333
31	Anatoly Ilyenkov	0.6265825307
32	Dmitriy Ukrainskiy	0.6278237636
33	Nikita Churkin	0.647591996
34	supertoha	0.6501102477
35	Sergei Dvindenko	0.6593224941
36	Альберт Тухватуллин	0.669298106
37	EkaterinaE	0.6766130021
38	Konstantin Phd	0.7378857027
39	Николай Брадис	0.7437817052
40	Etvas Никто	0.7557889485
41	Kirill Gerasimov	0.7728035492
42	Twins	0.7832226224
43	SG_da (Максим Сигида)	0.8183912617
44	Invincible	0.8466846174
45	Андрей Кузнецов	0.8475220699
46	Electric Boogaloo	0.8810139644
47	Nikolay Kobalo	0.9037624272
48	SpaceDoge	0.9593949661
49	MR M	1.0348883271
50	Максим Шульга	1.1221059082