Financial market analysis using time-series models, clustering algorithms, Transformers, and reinforcement learning for trading strategies.
Last updated Jun 16, 2026
10
Stars
0
Forks
0
Issues
0
Stars/day
Attention Score
39
Topics
Language breakdown
No language data available.
▸ Files
click to expand
README
name: Financial Market Analysis (Double-Tower Transformer) type: Quantitative Research / ML Engineering
Content
Intro - What is it? Briefly
A predictive machine learning system explicitly designed to generate profitable trading signals from high-frequency (1-minute) EUR/USD forex data. The flagship focus of this repository is a custom Double-Tower Gated Transformer Network (GTN) built to predict Take Profit / Stop Loss (TP/SL) strikes.How Does it Work? Technical Spec.
The data pipeline processes 1-minute OHLC bars to map out realistic TP/SL bounds over a 24-hour lookahead rather than using fixed time horizons. The core model, the Gated Transformer Network (GTN), utilizes a dual-tower architecture that separates feature interactions from temporal sequences. It replaces traditional sinusoidal encodings with learned time embeddings to capture market regime shifts and session overlaps (London, NY, Tokyo). To combat severe class imbalance (the model's tendency to constantly predict "keep/hold"), the network optimizes against a custom cost matrix and weighted cross-entropy loss.What Sets This Project Apart?
Instead of optimizing for standard classification accuracy, the pipeline tackles raw profitability. The models learn the structural mechanics of price action, balancing Risk/Reward ratios (e.g., 1:2) against real-world MetaTrader 5 execution constraints (enforcing next-candle open entries and proper UTF-16-LE signal formatting).Evals and test Series
- Primary Metric: Composite Score =
(buyprecision + sellprecision)/2 - 0.25 * |buyprecision - sellprecision| - Best GTN Score: 0.36 Composite
- Baseline (LightGBM): 0.37 Composite (but highly stochastic and prone to overfitting the 'keep' class).
- Status: The dual-tower architecture theoretically maps the feature-space better than tree-based models, but converting high composite scores to live P&L in MT5 remains the primary hurdle.
