Tests catalogue¶

This page indexes the test files in the tests/ directory by functional domain (current count = find tests -type f -name '*.py' | wc -l, growing as Stories land). Per project policy, every behavior worth shipping has a test, and every test worth keeping is discoverable here.

If a test isn't listed below, it's either (a) a temporary ad-hoc test (in which case it should be moved or deleted) or (b) the index is stale (please open a follow-up Story to fix it — see CVN-N011-EA).

How to run¶

make test-unit          # Fast unit tests
make test-integration   # Slow integration tests (external deps)
make test-cov           # With coverage
make qa                 # Format + lint + test-unit

Markers : @pytest.mark.unit, @pytest.mark.integration, @pytest.mark.cache, @pytest.mark.etl, @pytest.mark.ml.

Tests by mission area¶

🎯 ML Boost — Track 5 + 6 + Pipeline contract hardening ¶

File	Locks down
`tests/integration/test_track5_label_smoothing.py`	Track 5 variant matrix end-to-end through `trainer.train(...)` ; type preservation (DataFrame in → DataFrame out) ; calibration contract on `(X_val, y_val)` ; trainer-side `_assert_dmatrix_contract` + `_assert_calibration_targets_discrete` regression bars
`tests/integration/test_track6_focal_loss.py`	Track 6 focal loss 5-variant matrix + temperature scaling + joint Track 5 × Track 6 + XGBWrapper raw_output_logits + FTF factor registration + 6 guardrail rejection cases
`tests/unit/training/labels/test_label_pipeline.py`	Hamilton dataflow nodes (`smoothed_y`, `cv_pred_probs`, `suspect_mask`, `final_y`, `final_weights`) + `TestSuspectMaskPerClassCap` (S08 regression bar : 8 deterministic tests via monkeypatch on `cleanlab.filter.find_label_issues`)
`tests/unit/training/XGBoost/test_focal_loss_formal_verification.py`	Track 6 SymPy symbolic verification (32 tests : grad/hess equivalence at rel=1e-12 + finite-difference sanity + math invariants)
`tests/unit/test_calibration_metrics.py`	ECE / Brier computation primitives
`tests/unit/test_oos_calibrator.py`	OOS calibration aggregation per fold
`tests/unit/test_threshold_calibrator.py`	ThresholdCalibrator (per-fold theta from f1_binary curve)
`tests/unit/test_classification_metrics.py`	f1_buy, AUC-buy, action_rate primitives
`tests/unit/test_calibration_validation.py`	Pre-calibration data validation
`tests/unit/test_threshold_method_routing.py`	CVN_THRESHOLD_METHOD env routing
`tests/unit/test_hpo_objectives_f1_binary.py`	HPO objective = f1_binary plumbing
`tests/unit/test_model_trainers.py`	Trainer entry points (XGBoost)
`tests/unit/test_hyperoptimizer_warm_start.py`	HPO warm-start

🎯 Filter tuning — win ratio mission ¶

File	Locks down
`tests/unit/test_filter_chain_executor.py`	FilterChainExecutor — orchestration + ordering
`tests/unit/test_filter_contracts.py`	Plugin contract (each filter's input/output shape)
`tests/unit/test_filter_plugins.py`	Plugin registry + lookup
`tests/unit/test_filter_diagnostics.py`	Per-filter diagnostic counters (raw vs accepted)
`tests/unit/test_filter_gateway_service.py`	FilterGateway service-side
`tests/unit/test_filter_gateway_client.py`	FilterGateway client-side
`tests/unit/test_engine_filter_migration.py`	Migration from legacy filter pattern
`tests/unit/test_cooldown_plugin.py`	Cooldown filter
`tests/unit/test_confirmation_filter.py`	Confirmation filter (Option D 2-candle pattern)
`tests/unit/test_cascade_cooldown.py`	Cascade cooldown across symbols
`tests/unit/test_trade_quality_filter.py`	Quality filter (last-stage gate)
`tests/unit/test_meta_label.py`	Meta-label filter (López de Prado)
`tests/unit/test_meta_label_orchestrator.py`	Meta-label per-fold orchestration
`tests/unit/test_regime_tagger.py`	Regime tagging (used by regime filter)
`tests/unit/test_regime_detector.py`	Regime detection algorithm
`tests/unit/test_cost_model.py`	Cost filter (slippage + fees)

🎯 Economic value growth mission ¶

File	Locks down
`tests/unit/test_etv_scoring.py`	Economic Trade Value scoring per opportunity
`tests/unit/audit/test_economic_thresholds.py`	Economic break-even floors per crypto
`tests/unit/test_atr_labeling.py`	ATR-based labeling (used by PTE strategy)
`tests/unit/test_atr_stops.py`	ATR-based dynamic stops
`tests/unit/test_check_exit_continuous.py`	Continuous-exit logic
`tests/unit/test_p040_vol_stable.py`	P040 — atr_1h stable feature for PTE
`tests/unit/test_p038_build_once.py`	P038 — atr_normalized once-per-fold
`tests/unit/test_p033_candle_alignment.py`	Candle alignment for backtest fairness
`tests/unit/test_walk_forward_optimizer.py`	Walk-forward θ optimizer

Tests by infrastructure area¶

Backtest engine + WFRB¶

File	Locks down
`tests/unit/test_backtest_engine.py`	`cvntrade_backtest_engine` LdP + Legacy candle loops
`tests/unit/test_backtest_api.py`	Backtest API surface
`tests/unit/test_wf_backtest.py`	Walk-forward backtest aggregation
`tests/unit/test_wfrb_decision_folds.py`	WFRB decision per fold
`tests/unit/test_wfrb_mlflow_query.py`	WFRB MLflow result query

FTF (finetune) framework¶

File	Locks down
`tests/unit/test_finetune_matrix.py`	Ablation factor matrix definitions
`tests/unit/test_finetune_baselines.py`	Baseline env (`ftf_baseline.json` fallback chain)
`tests/unit/test_finetune_persistence.py`	`finetune_results` Postgres persistence
`tests/unit/test_finetune_report.py`	PDF report generator
`tests/unit/test_finetune_stats.py`	BH-corrected p-values + bootstrap CI + Cohen's d
`tests/unit/test_finetune_power_mode.py`	power_mode (HPO trial counts per crypto group)
`tests/unit/test_ftf_guardrails.py`	FTF env guardrails (label_smoothing, cleanlab, focal_loss, PTE envelope, n_features)
`tests/unit/test_ftf_config_diff.py`	Console diff display
`tests/unit/test_ftf_config_ui_save_restore.py`	Console save/restore round-trip
`tests/unit/test_dag_finetune_pte_config.py`	DAG `dag_finetune__pte` config + run_id parsing
`tests/unit/finetune/test_persist_predictions.py`	Predictions table population
`tests/unit/test_preflight_hamilton_exec.py`	FTF preflight Hamilton executor
`tests/unit/test_preflight_framework.py`	Preflight step framework
`tests/unit/test_preflight_fi_reference.py`	FI reference cache preflight

Pipeline (enrichment, FE, inference, runner)¶

File	Locks down
`tests/unit/pipeline/test_inference_api.py`	InferenceAPI signal generation + thresholding
`tests/unit/pipeline/test_pipeline_parity.py`	Batch ↔ streaming parity (per #614 spec)
`tests/unit/pipeline/test_fe_infinity_guard.py`	FE infinity / NaN guard
`tests/unit/pipeline/test_pipeline_runner.py`	CandlePipeline runner (high-level orchestration)
`tests/unit/pipeline/runner/test_assertion.py`	Runner assertion stack
`tests/unit/pipeline/runner/test_builder.py`	Pipeline builder
`tests/unit/pipeline/runner/test_candle_loop.py`	Candle loop primitives
`tests/unit/pipeline/runner/test_steps.py`	Step composition
`tests/unit/test_pipeline_candle_loop.py`	Legacy candle loop
`tests/unit/test_enrichment_service.py`	EnrichmentAPI service-side
`tests/unit/test_enrichment_client.py`	EnrichmentAPI client-side
`tests/unit/test_enrichment_serialization.py`	Enrichment artefact serialization
`tests/unit/test_fe_service.py`	FeatureEngineeringAPI service
`tests/unit/test_fe_client.py`	FeatureEngineeringAPI client
`tests/unit/test_fe_split_embargo.py`	Split + embargo applied to FE
`tests/unit/test_fe_reindex.py`	FE reindex on cache miss
`tests/unit/test_compatible_train_builder.py`	Compatible train builder (FE artefact pinning)
`tests/unit/test_compatibility_gate.py`	Per-fold compatibility gate

Feature engineering components (legacy module)¶

File	Locks down
`tests/dataprep/FE_composants/test_analyseur_stationnarite.py`	Stationarity analyzer
`tests/dataprep/FE_composants/test_debruiteur_donnees.py`	Wavelet denoiser
`tests/dataprep/FE_composants/test_enrichisseur_donnees.py`	Indicator enricher
`tests/dataprep/FE_composants/test_etiquetage_donnees.py`	Triple-barrier labeling
`tests/dataprep/FE_composants/test_filtre_colonnes_inutiles_dict.py`	Useless-column filter
`tests/unit/test_feature_selection.py`	Feature selection (Variance + IF)
`tests/unit/test_feature_selection_framework.py`	FS framework (CVN-N001-ED, ADR-67)
`tests/unit/test_feature_importance_cache.py`	FI reference cache
`tests/unit/test_guardrails_feature_selection_methods.py`	FS method registry mirror sync

CUSUM + purged k-fold¶

File	Locks down
`tests/unit/test_purged_kfold.py`	PurgedKFold splitter (López de Prado)
`tests/unit/test_cusum_filter_fit.py`	CUSUM σ fit on training cache
`tests/unit/test_cusum_manager.py`	CUSUM manager state
`tests/unit/test_cusum_incremental.py`	Streaming CUSUM update
`tests/unit/test_cusum_no_leakage.py`	CUSUM σ comes from training cache (no leakage)
`tests/unit/test_cusum_label_alignment.py`	CUSUM gating ↔ label alignment
`tests/unit/test_cusum_gate_warn.py`	CUSUM gate warning emission
`tests/unit/test_cusum_training_mode_disabled.py`	CUSUM training mode (rejected variant per Phase 1b)
`tests/unit/test_issue_295_cusum_before_split.py`	CUSUM applied before split (regression bar)
`tests/unit/test_issue_234_cusum_blocked.py`	Issue #234 — CUSUM blocked symbols

Cache + versioning¶

File	Locks down
`tests/test_cache_architecture.py`	3-layer cache (L1/L2/L3)
`tests/test_intelligent_cache.py`	Adaptive cache decisions
`tests/test_coherent_cache_naming.py`	Cache key naming convention
`tests/test_cache_loading_sprint6.py`	Sprint 6 cache loading
`tests/integration/test_cache_hierarchy.py`	Cache hierarchy fallback chain
`tests/unit/test_cache_manager.py`	CacheManager API
`tests/unit/test_cache_manager_version_propagation.py`	Version propagation across layers
`tests/unit/test_cache_manager_version_extract.py`	Version extract from artefact
`tests/unit/test_cache_versioning.py`	Versioning policy
`tests/unit/test_cache_configuration.py`	Cache config
`tests/unit/test_cache_interface_simple.py`	Simple cache interface
`tests/unit/test_fs_cache_drift_invalidation.py`	FS cache drift invalidation

Audit (CVN-N007 audit framework)¶

File	Locks down
`tests/unit/audit/test_artefacts.py`	Artefact integrity checks
`tests/unit/audit/test_baselines.py`	Baseline comparison
`tests/unit/audit/test_cli.py`	Audit CLI
`tests/unit/audit/test_economic_thresholds.py`	Economic break-even per crypto
`tests/unit/audit/test_label_noise.py`	Label noise detection
`tests/unit/audit/test_leakage.py`	Temporal leakage detection
`tests/unit/audit/test_predictions.py`	Predictions integrity
`tests/unit/audit/test_report.py`	Audit report rendering
`tests/unit/audit/test_signal_ceiling.py`	Signal ceiling diagnostic
`tests/unit/audit/test_statistical_floor.py`	Statistical floor diagnostic

Trading runtime¶

File	Locks down
`tests/unit/test_paper_trading_engine.py`	Paper trading engine
`tests/integration/test_live_trading_safety.py`	Live trading safety guards
`tests/unit/test_live_trading_config.py`	Live config
`tests/unit/test_live_feature_engine.py`	Live FE
`tests/unit/test_signal_generator.py`	Signal generator
`tests/unit/test_state_machine.py`	Position state machine
`tests/unit/test_state_manager.py`	State manager (event-sourced sessions)
`tests/unit/test_event_store.py`	Event store
`tests/unit/test_pt_session_isolation.py`	Paper trading session isolation
`tests/unit/test_order_reconciler.py`	Order reconciler
`tests/unit/test_circuit_breaker.py`	Circuit breaker (kill-switch foundation)
`tests/unit/test_risk_manager.py`	Risk manager
`tests/unit/test_promotion_workflow.py`	Champion/challenger promotion
`tests/unit/test_cross_env_promotion.py`	Cross-env promotion
`tests/unit/test_health_check.py`	Health check
`tests/unit/test_market_data_service.py`	Market data service
`tests/unit/test_market_data_client.py`	Market data client
`tests/unit/test_micro_api_base.py`	Micro-API base class
`tests/unit/test_scoring_client.py`	Scoring client (Inference façade)

MLflow + model registry¶

File	Locks down
`tests/unit/test_model_registry_service.py`	Model Registry service-side
`tests/unit/test_model_registry_client.py`	Model Registry client-side
`tests/unit/test_model_monitor.py`	Model monitoring
`tests/unit/test_lstm_predictor.py`	LSTM predictor (deferred Track 8)

Drift detection + monitoring¶

File	Locks down
`tests/unit/test_drift_detector.py`	Drift detection algorithm
`tests/unit/test_evidently_reporter.py`	Evidently report rendering
`tests/unit/test_mae_tracking.py`	MAE tracking
`tests/unit/test_shadow_first_divergence.py`	Shadow-mode first-divergence detection
`tests/unit/test_observability_otel.py`	OTel emission
`tests/unit/test_alerting.py`	Alert routing
`tests/unit/test_structured_logger.py`	Structured logging contract (ADR-30/32/33)
`tests/unit/test_dashboard_utils.py`	Dashboard helper utilities

Configuration + validators¶

File	Locks down
`tests/unit/test_base_config.py`	Base config hierarchy
`tests/unit/test_config_loader.py`	Config loader
`tests/unit/test_data_validators.py`	Data validators
`tests/unit/test_crypto_groups.py`	Crypto group definitions
`tests/unit/test_timeframe_guard.py`	Timeframe guardrail
`tests/unit/test_train_window_effective.py`	Effective training window calc
`tests/unit/test_timeout_per_crypto.py`	Per-crypto timeout
`tests/unit/test_strategy_testing.py`	Strategy testing framework
`tests/unit/test_strategy_grid_search.py`	Strategy grid search
`tests/unit/test_verdict.py`	Verdict (gate decision)

DAGs + launchers¶

File	Locks down
`tests/dags/test_dag_domain_architecture.py`	DAG domain architecture (`dag_{domain}__{action}`)
`tests/dags/test_launcher_dags.py`	Launcher DAGs (per ADR-19)

Pipelines + scoring (legacy module names)¶

File	Locks down
`tests/training/TwoStage/test_hyperoptimizer.py`	Two-stage HPO
`tests/test_hyperopt_restored.py`	HPO restoration
`tests/unit/regime/test_capture_oos_predictions.py`	Regime OOS predictions capture
`tests/unit/test_adaptive_event_engine.py`	Adaptive event engine
`tests/unit/test_pipeline_runner.py`	Pipeline runner
`tests/unit/test_gs_optimizations.py`	Grid search optimizations
`tests/unit/test_gs_artifact_store.py`	Grid search artefact store v1
`tests/unit/test_gs_artifact_store_v2.py`	Grid search artefact store v2
`tests/unit/test_gs_analyze.py`	Grid search analyze

Feast + feature store¶

File	Locks down
`tests/unit/test_feast_inference.py`	Feast inference path
`tests/unit/test_feast_config.py`	Feast config

Issue-specific regression bars (legacy naming)¶

These tests carry the issue number in their filename — they are regression bars for specific historical bugs.

File	Issue context
`tests/unit/test_issue_224_per_crypto_strategies.py`	Per-crypto strategy isolation
`tests/unit/test_issue_227_cost_filter_atr.py`	Cost filter ATR usage
`tests/unit/test_issue_234_cusum_blocked.py`	CUSUM blocked symbols
`tests/unit/test_issue_235_feature_mismatch.py`	Feature mismatch on restore
`tests/unit/test_issue_237_pt_events.py`	Paper trading event flow
`tests/unit/test_issue_254_backtest_execution.py`	Backtest execution semantics
`tests/unit/test_issue_259_pt_universe.py`	Paper trading universe
`tests/unit/test_issue_259_pt_allocation_gap.py`	Paper trading allocation gap
`tests/unit/test_issue_294_hpo_calibration.py`	HPO calibration ordering
`tests/unit/test_issue_295_cusum_before_split.py`	CUSUM before split

Misc¶

File	Locks down
`tests/test_secrets/test_secrets_manager.py`	Secrets manager
`tests/unit/test_openproject_import_gh.py`	OpenProject ↔ GH issue sync script
`tests/unit/test_stores.py`	Generic store utilities
`tests/unit/test_product_models.py`	Product models

Coverage policy¶

Per ADR-58, every FTF factor MUST have a guardrail + integration test ; the integration test goes under tests/integration/test_<track>_<factor>.py and locks down the variant matrix end-to-end through trainer.train(...) (see test_track5_label_smoothing.py and test_track6_focal_loss.py for the canonical pattern).

Per the CVN-N011-EA-S02 ADR amendment (in plan), every Story touching a pipeline boundary MUST have an integration test that replays the actual downstream consumer call (not a mock). This is the lesson from incidents §17.1-§17.3.