https://github.com/1adore1/deadlock-match-tracker-bot
Telegram bot for tracking real-time Deadlock matches for top 250 players of the leaderboard. Fetches match data and predicts winners using a machine learning model.
https://github.com/1adore1/deadlock-match-tracker-bot
aiogram api deadlock optuna pandas python scikit-learn
Last synced: 2 months ago
JSON representation
Telegram bot for tracking real-time Deadlock matches for top 250 players of the leaderboard. Fetches match data and predicts winners using a machine learning model.
- Host: GitHub
- URL: https://github.com/1adore1/deadlock-match-tracker-bot
- Owner: 1adore1
- Created: 2025-01-13T12:16:29.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-01-27T12:39:22.000Z (over 1 year ago)
- Last Synced: 2025-03-20T15:14:00.365Z (over 1 year ago)
- Topics: aiogram, api, deadlock, optuna, pandas, python, scikit-learn
- Language: Jupyter Notebook
- Homepage:
- Size: 422 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Deadlock Match Tracker Bot
## Description
Deadlock Match Tracker Bot is a Telegram bot designed to track and analyze matches in the game Deadlock for top 250 players of the leaderboard. It collects match data, processes it, and provides insights using a trained machine learning model.
## Features
- Fetch and store match data
- Process and clean match results
- Train a machine learning model for match predictions
- Provide match analysis via a Telegram bot
## Machine Learning Model
- Used **LightGBM** for match prediction
- Achieved **ROC AUC = 0.71**
- Used **Optuna** for hyperparameter tuning
## Project Structure
```
deadlock-match-tracker-bot/
│── 📂 assets/ # Media files
│ │── img.jpg # Example image
│── 📂 data/ # Datasets used in the project
│ │── clean_data.csv # Processed match data
│ │── match_results.csv # Raw match results
│── 📂 models/ # Trained machine learning models
│ │── model.joblib # Serialized model file
│── 📂 notebooks/ # Jupyter notebooks for data processing and model training
│ │── fetch_data.ipynb # Notebook for data collection
│ │── process_data.ipynb # Notebook for data preprocessing
│ │── train_data.ipynb # Notebook for model training
│── 📂 src/ # Source code of the project
│ │── tg_bot.py # Telegram bot implementation
│ │── tools.py # Utility functions
│── requirements.txt # List of dependencies
```
## Installation
1. Clone the repository:
```
git clone https://github.com/1adore1/deadlock-match-tracker-bot.git
```
2. Navigate to the project directory:
```
cd deadlock-match-tracker-bot
```
3. Install dependencies:
```
pip install -r requirements.txt
```
## Usage
1. Enter your telegram bot API_KEY in `src/config.py`.
2. Run the Telegram bot:
```
python src/tg_bot.py
```
3. Interact with the bot via Telegram to get match analytics and predictions.