artrdon/dquant
Библиотека для обучения моделей машинного обучения для предсказывания волатильности, не разбираясь в машинном обучении.
What's novel
Библиотека для обучения моделей машинного обучения для предсказывания волатильности, не разбираясь в машинном обучении.
Code Analysis
4 files read · 3 roundsA Python library that automates volatility forecasting for financial time series using gradient boosting models (XGBoost/LightGBM) with custom feature engineering and ONNX export capabilities.
Strengths
Provides a complete end-to-end pipeline from raw OHLCV data to model predictions, including robust visualization tools and support for multiple data sources like Yahoo Finance and MetaTrader 5. The ONNX export feature adds significant value for production deployment.
Weaknesses
Lacks any test suite, has poor naming conventions (cryptic variable names), mixes languages in comments (Russian/English), and contains potential syntax errors ('raise "pizdec"'). Feature engineering is basic and lacks the 'dozens of features' implied by the README.
Score Breakdown
Signal breakdown
Innovation
Craft
Traction
Scope
Evidence
Commits
27
Contributors
1
Files
11
Active weeks
3
Repository
Language
Python
Stars
1
Forks
0
License
MIT