dota2-predictor and dota2predictor
These are competitors offering alternative machine learning approaches to the same problem—one provides a general ML predictor while the other packages XGBoost predictions into a Telegram bot interface for different use cases.
About dota2-predictor
andreiapostoae/dota2-predictor
Tool that predicts the outcome of a Dota 2 game using Machine Learning
Extracts game data via the OpenDota API and mines match histories into structured datasets, then uses Logistic Regression and Neural Networks to compute hero synergy/counter matrices as engineered features. Supports both full 10-hero predictions and partial 9-hero queries that recommend optimal last picks while filtering for team composition compatibility using role-based similarity scoring, with preprocessing stratified by MMR brackets.
About dota2predictor
masterhood13/dota2predictor
Dota 2 Match Result Predictor Telegram Bot Overview This project is a Telegram bot that leverages a XGBoost neural network model to predict the outcomes of Dota 2 matches. The bot provides users with real-time predictions based on current match data, making it a useful tool for Dota 2 enthusiasts and analysts.
Scores updated daily from GitHub, PyPI, and npm data. How scores work