Top.Mail.Ru
Main page Schoolchildren Training tasks Hackaton ai-academy 2019. Online stage
HACKATON AI-ACADEMY
2019
ONLINE STAGE

Determine the player's experience in the Dota 2 game based on the statistics of one match.

@
WHAT IS HAPPENING IN DOTA 2?

The game Dota 2 involves two sides: teams of the forces of Light (The Radiant) and the forces of Darkness (The Dire). Each team consists of 5 players - characters with their own unique abilities.

The goal of the game is to destroy the throne located in the enemy base.

@
The task

Your task is to build an algorithm that can determine a player’s experience based on statistics from one match.

Training data for building models and test data for testing algorithms are given. The solution to the problem is the predictions of your algorithms on test data.

Data

To solve the problem, participants were given a set of almost 50,000 examples of matches in which the player’s level was known. For each of the ≈21,000 test cases, participants had to analyze the match data using their algorithms and answer whether the player was skilled.

Each data example provided describes the characteristics of the match and the statistics of one of the players at the end of the match. All examples have a unique identifier "id".

Two data sets were prepared:

  • Train, in which the player's experience medal type is known for each example.
  • Test, where the participants do not know the player’s experience - it needs to be calculated.

Data sets were provided in two formats:

  • CSV tables: skill_train.csv, skill_test.csv.
  • JSON documents: skill_train.jsonlines, skill_test.jsonlines (extended, contains additional information that is not in CSV tables).
Show Hide
Download Данные.zip
Solution Format

For each example from the test set, it is necessary to predict the player's experience. You need to provide the system with a CSV table with predictions for verification; it should contain two columns: id — example identifier, skilled — medal type (0 - beginner, 1 - experienced).

The quality of the solution was considered as the proportion of correctly guessed “experienced”/“novice” labels for all test examples. A detailed description of the metric is available at the link below.

To solve this, it is most convenient to use the Pythonprogramming language, since it has a large number of libraries for data analysis: NumPy, Pandas, SciKit-Learn and others. The Jupyter interactive environment is used as a development tool.

A basic example solution from the developers is available as a Jupyter notebook.

About AI Olympiad 2020-2021
@
Project organizer: Charitable Foundation “Investment to the Future”, OGRN 1157700017518
The Academy of Artificial Intelligence for Schoolchildren is not an educational service subject to licensing and does not imply the issuance of a state certificate
Our social networks
Do you have any questions or suggestions?
Contact
For schoolchildren, teachers and parents