Family
LSTM + XGBoost ensemble (per-pair)
in-developmentApache 2.0tabularboth
Forex Remittance Predictor.
Forex rate predictor (LSTM + XGBoost ensemble, per-pair).
Foundation · LSTM + XGBoost ensemble (per-pair)
What it does
The honest description.
Forex rate predictor (LSTM + XGBoost ensemble, per-pair).
Architecture
How it works.
Currency pair historical forex rates (date, currency_pair, OHLC), remittance volume signals, macro indicators. Per-pair training activates when input has 'currency_pair' column; falls back to 'default' single model otherwise. Per-pair forex rate predictions for remittance corridors (7d, 30d forecasts). LSTM + XGBoost ensemble weighted average. n_per_pair_models surfaced in metrics.
Training data & drift
What it learned from, and how we watch it.
Source
Inputs: Currency pair historical forex rates (date, currency_pair, OHLC), remittance volume signals, macro indicators. Per-pair training activates when input has 'currency_pair' column; falls back to 'default' single model otherwise.
Drift baseline · current quarter
Drift baseline not yet documented for this model.
What it's NOT
- PR #11 of model audit roadmap: trainable=True flipped, lifecycle=ACTIVE promoted, default_cost_per_run set.
Version history · immutable once published
Every release, traceable.
Data Mukodi holds about you is subject to DPDP §12 erasure (use /consent in the app), and any concern about this model can be raised via [email protected] under DPDP §13 grievance, or via /security.