Discord

Checklist ROAI 2025 County Stage · Task 1

Calorie Consumption

Romanian title: Consum Caloric

Compute four statistics of an activity dataset and predict calories burned, including for a subset representing a men's handball team.

  • Tabular
  • Descriptive statistics + regression
  • Romanian original · English translation
Download the English statement Read the official statement (Romanian)

The task

Each record describes a physical activity by user ID, gender, age, height, weight, duration, average heart rate and average body temperature; the training set also gives the calories burned.

Part 1 asks for the number of training rows, the number of rows for men, the mean activity duration and the number of users aged 75 or over. Part 2 asks for predicted calories for test rows with Subtask = 5 and for test rows with Subtask = 6 (a men's handball team).

Abridged and translated by SOTA from the official Romanian materials. The official statement has the exact rules, and it wins wherever this summary differs.

In English

This task was published in Romanian. SOTA translated it into English on 16 September 2026.

Read the task statement in English 457 words

Calorie Consumption

English translation by SOTA – AI Community of the Romanian original, published with the permission of Asociația Nitro. Organisers who would like this translation removed can email [email protected].

Statement

Implement an AI/ML model to predict the calorie consumption of an activity, given a training set train_data.csv and the test set test_data.csv (on which you only need to make predictions).

The dataset contains the following fields, with these meanings:

  • User_ID: the ID of the user who performed the activity
  • Gender: the user's gender
  • Age: the user's age
  • Height: the user's height in cm
  • Weight: the user's weight in kg
  • Duration: the duration of the activity
  • Heart_Rate: the average heart rate during the activity
  • Body_Temp: the average temperature during the activity
  • Calories: the number of calories consumed during the activity (present only in the training dataset)

Notes on the dataset:

The target attribute is Calories. Given the other attributes, the goal is to predict Calories for the test dataset. The evaluation metric used is MAE (the mean of the absolute differences between the value predicted by the algorithm and the correct value).

Requirements

Part 1 (20p). Based on the training dataset, compute the following:

"Samples, No.Males, AverageDuration, SeniorUsers"

  • Subtask 1: Samples: the number of rows in the training dataset
  • Subtask 2: No.Males: the number of training examples that describe activities performed by men
  • Subtask 3: AverageDuration: the average duration of the activities in the training set.
  • Subtask 4: SeniorUsers: the number of users in the training set who are at least 75 years old

Each of the 4 subtasks is worth 5p.

Part 2 (80p)

  • Subtask 5: (60p) Develop an AI/ML model and predict the Calories attribute for each example in the test data file for which the Subtask column has the value 5.

  • Subtask 6: (20p) A men's handball team needs to estimate calorie consumption in order to optimise the players' diet. Develop an AI/ML model and predict the number of calories for each example in the test data file for which the Subtask column has the value 6.

Output format

A csv file output.csv that includes the following 3 columns:

  • subtaskID - the number of the subtask (1, 2, 3, 4, 5 or 6)
  • datapointID - which refers to the User_ID column in test_data.csv
  • answer - the answer corresponding to the datapoint for the respective subtask

Note: For subtasks 1, 2, 3 and 4, which require a single answer for the whole dataset, output a single row whose datapointID is 1.

Submit a single csv containing the answers for all the subtasks you have solved. To see an example, download the file sample_output.csv

Translated by SOTA. The Romanian original is the official version and wins wherever the two differ. The statement gives 5 points to each of Subtasks 1–4; the judge's own subtask maxima are 3, 4, 6 and 7 (still 20 in total). Published with the permission of Asociația Nitro, which runs ROAI (17 September 2026). If you organise this olympiad and would like the translation removed, email [email protected] and we will take it down.

At a glance

You get
train_data.csv (9,000 rows: User_ID, Age, Height, Weight, Duration, Heart_Rate, Body_Temp, Gender, Calories) and test_data.csv (2,500 rows with a Subtask column: 2,000 rows for subtask 5 and 500 for subtask 6).
You submit
output.csv with subtaskID (1–6), datapointID (User_ID; 1 for subtasks 1–4) and answer.
Scoring
Subtasks 1–4: exact answers (5 points each in the statement). Subtask 5 (60 points) and subtask 6 (20 points): MAE-based scoring; thresholds are not given in the statement.
Rules
  • On-site at county competition centres; identity document and signed declaration required
  • Only submissions selected during the round count; anti-fraud session whitelist on the judge
Format
ROAI/ONIA 2025 county stage, 29 April 2025 (06:20–10:35 UTC on the judge); same problem for grades 9–10 and 11–12.

In English (SOTA)

Statement

Competitions

Solution

Source files

Details

Year
2025, County centres, Romania
Round
County Stage · Task 1
Language
Romanian; English translation by SOTA
License
Not stated by the source