Home

ELVIS Trading Bot - Project Overview

Introduction

The ELVIS Trading Bot is a modular, machine learning-driven algorithmic trading system for automated cryptocurrency trading. The system integrates multiple ML architectures (Random Forest, Neural Network, Transformer, Ensemble) and supports real-time data processing, risk management, and trade execution.

Key Features:

• Modular design with a clear separation of models, strategies, executors, and data pipelines
• Strategy pattern for flexible trading logic
• Comprehensive monitoring and visualization (Console Dashboard, Telegram Bot, Grafana)
• Backtesting and evaluation framework
• Secure configuration, secrets management, and extensive logging

System Architecture

The system is organized into the following core modules:

• Models: ML models for prediction (RandomForest, Neural Network, Transformer, Ensemble)
• Trading System: Strategies and execution backends
• Data Pipeline: Downloading, cleaning, and feature engineering on market data
• Training Pipeline: Distributed training for ML and RL models
• Utilities & Monitoring: Console dashboard, Telegram alerts, Prometheus/Grafana observability

See detailed mermaid diagrams and architecture docs in the /docs folder of the repository.

Project Structure (Excerpt)

BTC_BOT/
├── main.py
├── core/
│   ├── models/
│   ├── data/
│   ├── metrics/
├── trading/
│   ├── strategies/
│   ├── execution/
│   ├── data/
│   └── risk/
├── training/
│   ├── models/
│   └── training_results/
├── utils/
├── scripts/
├── models/
├── docs/
├── tests/
├── images/

Refer to the latest project_cleanup_summary.md for a complete structure and organization.

---

See Also

• Installation
• Usage
• FAQ