--expiry

🦉 AthenaAI is an AI-powered ensemble machine-learning trading bot for PocketOption, built on BinaryOptionsToolsV2 → Get it on GitLab
💬 Need help? Join the Chipa Discord · ⚡ Go no-code with ChipaEditor

The trade duration (in seconds) each training sample is labeled for. Train for the expiry you actually intend to trade.

What it does

Together with --timeframe, expiry determines the label horizon in Trainer.run():

label_horizon = max(1, int(cfg.expiry / max(cfg.timeframe, 1)))

The trainer logs it explicitly: Generating samples (label horizon = 5 candles = 300s) …. In build_samples() (src/training/dataset.py), each candle position i becomes a sample labeled 1 (CALL wins) if candles[i + label_horizon].close is above candles[i].close, 0 (PUT wins) if below, and is skipped entirely if the close is unchanged (draws are not trained on). This mirrors exactly how a binary option of that duration would settle.

Choosing a value

• Match your live trading expiry — e.g. PO_DEFAULT_EXPIRY=120 means you should train with --expiry 120. A brain labeled for 5-minute outcomes is not optimal for 1-minute trades; price behavior at different horizons differs.
• Values below --timeframe clamp to a 1-candle horizon; use finer candles for shorter expiries.
• Non-multiples of the timeframe truncate: --expiry 90 --timeframe 60 gives a 1-candle (60 s) horizon, not 90 s.
• Larger expiries consume more data per sample: the final label_horizon candles yield no samples, and adjacent samples’ labels overlap more, which makes the effective information per sample lower.

Risk note: the labeled expiry directly shapes what the brain trades live. A mismatch between the trained expiry and the live one degrades the edge; binary options trading can lose money and past performance doesn’t guarantee future results.

Examples

python train.py --expiry 120                      # label for 2-minute trades
python train.py --data EURUSD_M1.csv --expiry 60  # 1-minute trades on M1 (horizon = 1)

Interactions

• --timeframe — divisor in the horizon formula.
• PO_DEFAULT_EXPIRY and PO_EXPIRY_OPTIONS — the live expiries this training should match.

See also

• Expiry selection concept
• Training guide

💬 Unsure how this interacts with the rest of the configuration? Ask in the Chipa Discord — or prototype strategies no-code with ChipaEditor.