limen.targets

Build supervised target columns from market data for manifest-driven and custom experiments.

Canonical docs

• Targets
• Experiment Manifest
• Transforms

What this package owns

Owns train-fitted and transform-time target builders, including binary classification targets, return targets, breakout targets, barrier targets, and identity passthrough targets. Does not own feature engineering, model fitting, calibration, or execution strategy.

Key entry points

Entry point	Use case	Notes
QuantileBinaryTarget	Labels from a train-fitted quantile cutoff	Fits the cutoff on train only.
ThresholdBinaryTarget	Labels from a fixed threshold	Supports explicit return thresholds.
TradelineLongBinaryTarget	Long-entry label from price-line context	Pairs with line-based feature helpers.
NextReturnTarget, VolNormalizedReturnTarget, ForwardVolNormalizedReturnTarget	Continuous forward-return style targets	Common for regression or ranking experiments.
NextBarUpTarget, NextBarDownTarget	One-bar direction labels	Classification baselines.
ForwardBreakoutTarget, EmaBreakoutTarget	Breakout-oriented labels	Uses forward path or EMA context.
TripleBarrierTarget	Path-aware barrier label	Requires enough rows for max_horizon and the volatility warmup.
RiskRewardRatioTarget, ExitQualityTarget	Outcome labels from upstream trade-path columns	Require the expected input columns to exist on every split.
RandomBinaryTarget, IdentityTarget	Controls or already-present target columns	Supports tests, baselines, and externally labelled data.

Adjacent modules

• limen.transforms can derive target inputs before target construction.
• limen.experiment.Manifest wires target builders into prepare_data().
• limen.calibration evaluates predictions against the target labels after model output.

Quick orientation

targets/
|-- quantile_binary.py
|-- threshold_binary.py
|-- next_bar_up.py / next_bar_down.py
|-- next_return.py
|-- vol_normalized_return.py / forward_vol_normalized_return.py
|-- forward_breakout.py / ema_breakout.py / tradeline_long_binary.py
|-- triple_barrier.py
|-- risk_reward_ratio.py / exit_quality.py
|-- identity.py
`-- random_binary.py

Things to know

• Target classes receive the training split during construction and apply transform() to every split.
• Train-fitted targets must store only train-derived state before validation and test transformation.
• Forward-looking targets naturally create boundary nulls; manifest preparation drops null target rows downstream.
• Outcome-derived targets require their upstream columns to exist before target construction.

Read next

• Targets
• Experiment Manifest