stefan-jansen/machine-learning-for-trading

Pattern Findings

The scanner identified 15558 distinct pattern matches across 24 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a detector match to inspect in context. Expand any category row to inspect the individual file paths, line numbers, code snippets, and the lexical context (CODE, COMMENT, or STRING) in which each match was detected.

Reading the findings table: The Severity column indicates the configured finding weight (CRITICAL / HIGH / MEDIUM / LOW). The Context column identifies whether the match occurred inside executable code, an inline comment, or a string literal — legacy comment-context matches receive a ×1.5 score weight. The ⚡ bolt icon marks clustered matches: three or more patterns within a 10-line window, each receiving an additional ×1.5 density multiplier under the legacy scoring model.

Over-Commented Block6727 hits · 6311 pts

Severity	File	Line	Snippet	Context
LOW	.pre-commit-config.yaml	41		COMMENT
LOW	.pre-commit-config.yaml	61	- id: strip-plotly-json	COMMENT
LOW	pyproject.toml	21	"pyarrow>=15.0",	COMMENT
LOW	pyproject.toml	101	# ===================	COMMENT
LOW	pyproject.toml	261	# ruff versions format the same code differently, so a notebook formatted by	COMMENT
LOW	sitecustomize.py	161	configured = _data_root_from_dotenv(repo_root)	COMMENT
LOW	docker-compose.yml	1	# ML4T 3rd Edition — Docker Compose Configuration	COMMENT
LOW	docker-compose.yml	21	#	COMMENT
LOW	docker-compose.yml	41	# local_files_only=True, so a container run has to see the host's cache or the	COMMENT
LOW	21_rl_execution_hedging/01_algorithms_comparison.py	1	# ---	COMMENT
LOW	21_rl_execution_hedging/01_algorithms_comparison.py	21	# ## Purpose	COMMENT
LOW	21_rl_execution_hedging/01_algorithms_comparison.py	41	# - Train a value-based agent (DQN) and two actor-critic agents (PPO, A2C) with	COMMENT
LOW	21_rl_execution_hedging/01_algorithms_comparison.py	101		COMMENT
LOW	21_rl_execution_hedging/01_algorithms_comparison.py	141	# GARCH(1,1) models the variance of a return series as a function of its own	COMMENT
LOW	21_rl_execution_hedging/01_algorithms_comparison.py	301		COMMENT
LOW	21_rl_execution_hedging/01_algorithms_comparison.py	321	# sees the return $r_t$ that has just happened. The reward it collects is	COMMENT
LOW	21_rl_execution_hedging/01_algorithms_comparison.py	401	obs = build_observation(self.returns, self.step_idx, self.position)	COMMENT
LOW	21_rl_execution_hedging/01_algorithms_comparison.py	421	# means and why it needs fewer fresh environment steps per unit of progress.	COMMENT
LOW	21_rl_execution_hedging/01_algorithms_comparison.py	701	}	COMMENT
LOW	21_rl_execution_hedging/07_backtest_with_impact.py	1	# ---	COMMENT
LOW	21_rl_execution_hedging/07_backtest_with_impact.py	21	# ## Purpose	COMMENT
LOW	21_rl_execution_hedging/07_backtest_with_impact.py	41	# - Compute a name's participation rate - the size of an order against the	COMMENT
LOW	21_rl_execution_hedging/07_backtest_with_impact.py	101	LIQUID_MAX_PARTICIPATION = 0.10 # order below this share of daily volume is the liquid cohort	COMMENT
LOW	21_rl_execution_hedging/07_backtest_with_impact.py	321	# ## The Square-Root Impact Model	COMMENT
LOW	21_rl_execution_hedging/07_backtest_with_impact.py	481	summary	COMMENT
LOW	21_rl_execution_hedging/07_backtest_with_impact.py	561	after selection and does not feed back into it.	COMMENT
LOW	21_rl_execution_hedging/02_optimal_execution_ppo.py	1	# ---	COMMENT
LOW	21_rl_execution_hedging/02_optimal_execution_ppo.py	21	# ## Purpose	COMMENT
LOW	21_rl_execution_hedging/02_optimal_execution_ppo.py	41	# basis points of the arrival notional, and say which part of that cost comes	COMMENT
LOW	21_rl_execution_hedging/02_optimal_execution_ppo.py	141	# %% [markdown]	COMMENT
LOW	21_rl_execution_hedging/02_optimal_execution_ppo.py	221	)	COMMENT
LOW	21_rl_execution_hedging/02_optimal_execution_ppo.py	241	# depth falls. The simulator therefore makes a stressed regime a wider, deeper	COMMENT
LOW	21_rl_execution_hedging/02_optimal_execution_ppo.py	261	stressed state and the rest to the normal state.	COMMENT
LOW	21_rl_execution_hedging/02_optimal_execution_ppo.py	281	# walks down the available bids, and the size of that effect grows with the	COMMENT
LOW	21_rl_execution_hedging/02_optimal_execution_ppo.py	301	# The action is a pace multiplier rather than a quantity so that no policy can	COMMENT
LOW	21_rl_execution_hedging/02_optimal_execution_ppo.py	401	)	COMMENT
LOW	21_rl_execution_hedging/02_optimal_execution_ppo.py	461	#	COMMENT
LOW	21_rl_execution_hedging/03_market_making_ppo.py	1	# ---	COMMENT
LOW	21_rl_execution_hedging/03_market_making_ppo.py	21	# ## Purpose	COMMENT
LOW	21_rl_execution_hedging/03_market_making_ppo.py	41	# After working through this notebook you will be able to:	COMMENT
LOW	21_rl_execution_hedging/03_market_making_ppo.py	141	a half-spread that already widens with volatility.	COMMENT
LOW	21_rl_execution_hedging/03_market_making_ppo.py	201	# ### What the maker sees and does	COMMENT
LOW	21_rl_execution_hedging/03_market_making_ppo.py	221	# notebook quote symmetrically around exactly this price, at three different	COMMENT
LOW	21_rl_execution_hedging/03_market_making_ppo.py	241	# $$r_t = \Delta W_t - \lambda_q \left(\frac{q_t}{q_{\max}}\right)^2 p_{t+1}$$	COMMENT
LOW	21_rl_execution_hedging/03_market_making_ppo.py	321	#	COMMENT
LOW	21_rl_execution_hedging/03_market_making_ppo.py	741	# ## 7. Which part of the inventory response is the policy's	COMMENT
LOW	21_rl_execution_hedging/04_crypto_execution_rl.py	1	# ---	COMMENT
LOW	21_rl_execution_hedging/04_crypto_execution_rl.py	21	# ## Purpose	COMMENT
LOW	21_rl_execution_hedging/04_crypto_execution_rl.py	41	# drawn from bars that had closed when the decision was made, and say which	COMMENT
LOW	21_rl_execution_hedging/04_crypto_execution_rl.py	141	)	COMMENT
LOW	21_rl_execution_hedging/04_crypto_execution_rl.py	181	# %%	COMMENT
LOW	21_rl_execution_hedging/04_crypto_execution_rl.py	321	)	COMMENT
LOW	21_rl_execution_hedging/04_crypto_execution_rl.py	341	#	COMMENT
LOW	21_rl_execution_hedging/04_crypto_execution_rl.py	721	# The flag fires when a policy reaches the final hour still holding more than it	COMMENT
LOW	21_rl_execution_hedging/04_crypto_execution_rl.py	761	# %% [markdown]	COMMENT
LOW	21_rl_execution_hedging/05_deep_hedging_pfhedge.py	1	# ---	COMMENT
LOW	21_rl_execution_hedging/05_deep_hedging_pfhedge.py	21	# ## Purpose	COMMENT
LOW	21_rl_execution_hedging/05_deep_hedging_pfhedge.py	41	# After working through this notebook you will be able to:	COMMENT
LOW	21_rl_execution_hedging/05_deep_hedging_pfhedge.py	61	#	COMMENT
LOW	21_rl_execution_hedging/05_deep_hedging_pfhedge.py	141	# expressed the same way and P&L is scaled back to currency at the end.	COMMENT
6667 more matches not shown…

Hyper-Verbose Identifiers5124 hits · 5248 pts

Severity	File	Line	Snippet	Context
LOW	21_rl_execution_hedging/02_optimal_execution_ppo.py	476	def discrete_almgren_chriss_schedule(	CODE
LOW	21_rl_execution_hedging/market_making_env.py	24	def generate_garch_market_data(	CODE
LOW	…execution_hedging/06_inverse_reinforcement_learning.py	205	def collect_expert_trajectories(	CODE
LOW	…execution_hedging/06_inverse_reinforcement_learning.py	565	def compute_feature_expectations(trajectories: list, extract_fn) -> np.ndarray:	CODE
LOW	…execution_hedging/06_inverse_reinforcement_learning.py	630	def sample_candidate_trajectory(env: ExecutionEnv, rng) -> dict:	CODE
LOW	…execution_hedging/06_inverse_reinforcement_learning.py	652	def trajectory_feature_counts(trajectories: list, extract_features_fn) -> np.ndarray:	CODE
LOW	…execution_hedging/06_inverse_reinforcement_learning.py	733	def regularized_log_likelihood(	CODE
LOW	…execution_hedging/06_inverse_reinforcement_learning.py	835	def sample_feature_matching_trajectory(	CODE
LOW	…execution_hedging/06_inverse_reinforcement_learning.py	1030	def inferred_boltzmann_action(	CODE
LOW	05_synthetic_data/00_classical_simulation.py	460	def simulate_mean_reversion_euler(	CODE
LOW	05_synthetic_data/00_classical_simulation.py	518	def simulate_mean_reversion_exact(	CODE
LOW	05_synthetic_data/04_gtgan_irregular.py	252	def generate_synthetic_irregular_bars(n_bars: int = 1000, seed: int = 42) -> pl.DataFrame:	CODE
LOW	05_synthetic_data/04_gtgan_irregular.py	352	def compute_inter_arrival_times(df: pl.DataFrame) -> np.ndarray:	CODE
LOW	05_synthetic_data/04_gtgan_irregular.py	383	def create_irregular_sequences(	CODE
LOW	05_synthetic_data/timegan_metrics.py	192	def compute_predictive_score_yoon(	CODE
LOW	05_synthetic_data/timegan_metrics.py	304	def compute_discriminative_score_yoon(	CODE
LOW	05_synthetic_data/timegan_metrics.py	555	def compute_discriminative_score(	CODE
LOW	05_synthetic_data/03_sigcwgan_signatures.py	651	def compute_signature_with_factorial_norm(	CODE
LOW	05_synthetic_data/03_sigcwgan_signatures.py	700	def augment_and_signature_paper(	CODE
LOW	05_synthetic_data/03_sigcwgan_signatures.py	810	def compute_signatures_batch_paper(	CODE
LOW	05_synthetic_data/03_sigcwgan_signatures.py	1186	def compute_expected_signature_gpu(	CODE
LOW	05_synthetic_data/03_sigcwgan_signatures.py	1666	def plot_path_comparison_unconditional(	CODE
LOW	05_synthetic_data/03_sigcwgan_signatures.py	1802	def tstr_evaluation_unconditional(	CODE
LOW	05_synthetic_data/02_tailgan_tail_risk.py	286	def deterministic_neural_sort(s: torch.Tensor, tau: float) -> torch.Tensor:	CODE
LOW	09_model_based_features/12_wasserstein_regimes.py	501	def simulate_two_regime_stream(	CODE
LOW	09_model_based_features/02_structural_breaks.py	743	def distribution_shift_features(series: np.ndarray, boundary: int, n_bins: int = 50) -> dict:	CODE
LOW	09_model_based_features/02_structural_breaks.py	778	def dependence_shift_features(series: np.ndarray, boundary: int) -> dict:	CODE
LOW	09_model_based_features/02_structural_breaks.py	849	def compute_all_break_features(series: np.ndarray, boundary: int) -> dict:	CODE
LOW	09_model_based_features/13_regime_as_feature.py	289	def evaluate_mixture_of_experts(estimator, features: list[str]) -> pd.DataFrame:	CODE
LOW	09_model_based_features/10_uncertainty_features.py	195	def fit_stochastic_volatility(observations: np.ndarray):	CODE
LOW	09_model_based_features/07_arima_features.py	348	def fit_recording_convergence(series: pd.Series, order: tuple[int, int, int]):	CODE
LOW	22_rag_financial_research/03_hybrid_retrieval.py	507	def semantic_similarity_scores(query: str, model, embeddings: np.ndarray) -> np.ndarray:	CODE
LOW	…_financial_research/02_domain_embeddings_comparison.py	351	def get_sentence_transformer_embeddings(	CODE
LOW	…_financial_research/07_institutional_holdings_graph.py	329	def compute_coownership_matrix(holdings: pl.DataFrame) -> tuple[np.ndarray, list[str]]:	CODE
LOW	…_financial_research/07_institutional_holdings_graph.py	729	def compute_institution_similarity(holdings: pl.DataFrame) -> pl.DataFrame:	CODE
LOW	12_gradient_boosting/11_conformal_gbm.py	110	def embargo_steps_from_buffer(label_buffer: str, dates: np.ndarray) -> int:	CODE
LOW	12_gradient_boosting/11_conformal_gbm.py	136	def chronological_calibration_masks(	CODE
LOW	12_gradient_boosting/11_conformal_gbm.py	161	def conformal_order_statistic(scores: np.ndarray, alpha: float) -> float:	CODE
LOW	12_gradient_boosting/11_conformal_gbm.py	218	def deterministic_lgb_parameters(objective: str, alpha: float \| None = None) -> dict:	CODE
LOW	12_gradient_boosting/11_conformal_gbm.py	503	def summarize_conformal_results(results: list[dict], alpha: float) -> dict \| None:	CODE
LOW	12_gradient_boosting/11_conformal_gbm.py	779	def compute_adaptive_intervals(	CODE
LOW	12_gradient_boosting/07_hpo_comparison.py	291	def optuna_continuous_objective(trial):	CODE
LOW	12_gradient_boosting/12_case_study_insights.py	1624	def pairwise_rank_correlation(fold_arrays: list[np.ndarray]) -> tuple[float, float] \| None:	CODE
LOW	02_financial_data_universe/06_futures_continuous.py	479	def identify_front_month_calendar(	CODE
LOW	…financial_data_universe/14_point_in_time_validation.py	274	def leakage_corr_scale_invariant(df: pl.DataFrame, feature: str, price: str = "close") -> float:	CODE
LOW	…financial_data_universe/14_point_in_time_validation.py	327	def validate_signal_trade_lag(signals: pl.DataFrame, execution_lag: int = 1) -> pl.DataFrame:	CODE
LOW	02_financial_data_universe/09_options_continuous.py	117	def select_constant_maturity_straddle(raw_options: pl.DataFrame) -> pl.DataFrame:	CODE
LOW	02_financial_data_universe/09_options_continuous.py	737	def build_continuous_straddle_series(cm_series: pl.DataFrame, held: pl.DataFrame) -> pl.DataFrame:	CODE
LOW	02_financial_data_universe/17_complete_pipeline.py	192	def _combine_pipeline_sources(	CODE
LOW	02_financial_data_universe/17_complete_pipeline.py	561	def _validate_crypto_coverage(df: pl.DataFrame, symbol: str) -> dict[str, Any]:	CODE
LOW	02_financial_data_universe/17_complete_pipeline.py	604	def _add_crypto_session_features(df: pl.DataFrame) -> pl.DataFrame:	CODE
LOW	…ancial_data_universe/15_survivorship_bias_detection.py	1073	def check_universe_completeness(	CODE
LOW	02_financial_data_universe/20_storage_benchmark_file.py	213	def read_parquet_materialized() -> pl.DataFrame:	CODE
LOW	02_financial_data_universe/20_storage_benchmark_file.py	250	def read_feather_materialized() -> pl.DataFrame:	CODE
LOW	02_financial_data_universe/02_corporate_actions.py	446	def demonstrate_split_adjustment(	CODE
LOW	02_financial_data_universe/02_corporate_actions.py	475	def demonstrate_dividend_adjustment(price_before_ex: float = 100.0, dividend: float = 5.0) -> None:	CODE
LOW	02_financial_data_universe/16_provider_comparison.py	560	def compare_providers_detailed(	CODE
LOW	16_strategy_simulation/08_signal_method_comparison.py	164	def select_registered_prediction(db_path: Path, family: str, label: str) -> dict[str, object]:	STRING
LOW	16_strategy_simulation/08_signal_method_comparison.py	239	def load_registered_predictions(	CODE
LOW	16_strategy_simulation/02_futures_backtesting.py	114	def load_contract_specs_from_yaml(yaml_path: Path) -> dict[str, ContractSpec]:	CODE
5064 more matches not shown…

Decorative Section Separators1171 hits · 3809 pts

Severity	File	Line	Snippet	Context
MEDIUM⚡	pyproject.toml	11	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	13	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	22	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	24	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	99	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	101	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	110	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	112	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	117	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	119	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	124	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	126	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	138	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	140	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	143	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	145	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	149	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	151	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	153	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	155	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	170	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	172	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	181	# ===================	COMMENT
MEDIUM⚡	pyproject.toml	183	# ===================	COMMENT
MEDIUM	pyproject.toml	37	# ===================	COMMENT
MEDIUM	pyproject.toml	39	# ===================	COMMENT
MEDIUM	pyproject.toml	53	# ===================	COMMENT
MEDIUM	pyproject.toml	55	# ===================	COMMENT
MEDIUM	pyproject.toml	70	# ===================	COMMENT
MEDIUM	pyproject.toml	72	# ===================	COMMENT
MEDIUM	pyproject.toml	85	# ===================	COMMENT
MEDIUM	pyproject.toml	87	# ===================	COMMENT
MEDIUM	docker-compose.yml	77	# ============================================	COMMENT
MEDIUM	docker-compose.yml	80	# ============================================	COMMENT
MEDIUM	docker-compose.yml	92	# ============================================	COMMENT
MEDIUM	docker-compose.yml	95	# ============================================	COMMENT
MEDIUM	docker-compose.yml	115	# ============================================	COMMENT
MEDIUM	docker-compose.yml	120	# ============================================	COMMENT
MEDIUM	docker-compose.yml	171	# ============================================	COMMENT
MEDIUM	docker-compose.yml	175	# ============================================	COMMENT
MEDIUM	docker-compose.yml	211	# ============================================	COMMENT
MEDIUM	docker-compose.yml	216	# ============================================	COMMENT
MEDIUM	docker-compose.yml	236	# ============================================	COMMENT
MEDIUM	docker-compose.yml	250	# ============================================	COMMENT
MEDIUM	docker-compose.yml	284	# ============================================	COMMENT
MEDIUM	docker-compose.yml	287	# ============================================	COMMENT
MEDIUM	docker-compose.yml	306	# ============================================	COMMENT
MEDIUM	docker-compose.yml	309	# ============================================	COMMENT
MEDIUM	05_synthetic_data/03_sigcwgan_signatures.py	852	# =============================================================================	COMMENT
MEDIUM	05_synthetic_data/03_sigcwgan_signatures.py	854	# =============================================================================	COMMENT
MEDIUM	05_synthetic_data/02_tailgan_tail_risk.py	717	# ---------------------	COMMENT
MEDIUM	05_synthetic_data/02_tailgan_tail_risk.py	719	# ---------------------	COMMENT
MEDIUM	05_synthetic_data/02_tailgan_tail_risk.py	739	# ---------------------	COMMENT
MEDIUM	05_synthetic_data/02_tailgan_tail_risk.py	741	# ---------------------	COMMENT
MEDIUM	13_dl_time_series/dl_sequences.py	42	# =============================================================================	COMMENT
MEDIUM	13_dl_time_series/dl_sequences.py	44	# =============================================================================	COMMENT
MEDIUM	13_dl_time_series/dl_sequences.py	113	# =============================================================================	COMMENT
MEDIUM	13_dl_time_series/dl_sequences.py	115	# =============================================================================	COMMENT
MEDIUM⚡	13_dl_time_series/dl_sequences.py	153	# =============================================================================	COMMENT
MEDIUM⚡	13_dl_time_series/dl_sequences.py	155	# =============================================================================	COMMENT
1111 more matches not shown…

Unused Imports995 hits · 864 pts

Severity	File	Line	Context
LOW	21_rl_execution_hedging/01_algorithms_comparison.py	88	CODE
LOW	21_rl_execution_hedging/07_backtest_with_impact.py	80	CODE
LOW	21_rl_execution_hedging/02_optimal_execution_ppo.py	95	CODE
LOW	21_rl_execution_hedging/market_making_env.py	3	CODE
LOW	21_rl_execution_hedging/rl_environments.py	14	CODE
LOW	21_rl_execution_hedging/03_market_making_ppo.py	95	CODE
LOW	21_rl_execution_hedging/04_crypto_execution_rl.py	91	CODE
LOW	21_rl_execution_hedging/rl_calibration.py	25	CODE
LOW	21_rl_execution_hedging/05_deep_hedging_pfhedge.py	103	CODE
LOW	…execution_hedging/06_inverse_reinforcement_learning.py	89	CODE
LOW	21_rl_execution_hedging/crypto_execution_env.py	3	CODE
LOW	05_synthetic_data/04_gtgan_irregular.py	138	CODE
LOW	05_synthetic_data/04_gtgan_irregular.py	140	CODE
LOW	05_synthetic_data/04_gtgan_irregular.py	142	CODE
LOW	05_synthetic_data/timegan_metrics.py	35	CODE
LOW	05_synthetic_data/05_diffusion_ts.py	105	CODE
LOW	05_synthetic_data/06_llm_tabular_great.py	84	CODE
LOW	05_synthetic_data/03_sigcwgan_signatures.py	148	CODE
LOW	05_synthetic_data/03_sigcwgan_signatures.py	150	CODE
LOW	05_synthetic_data/07_dp_gan.py	93	CODE
LOW	05_synthetic_data/02_tailgan_tail_risk.py	60	CODE
LOW	05_synthetic_data/02_tailgan_tail_risk.py	67	CODE
LOW	05_synthetic_data/01_timegan.py	83	CODE
LOW	09_model_based_features/09_har_rough_volatility.py	62	CODE
LOW	09_model_based_features/13_regime_as_feature.py	64	CODE
LOW	09_model_based_features/10_uncertainty_features.py	56	CODE
LOW	22_rag_financial_research/03_hybrid_retrieval.py	84	CODE
LOW	22_rag_financial_research/01_sec_filing_pipeline.py	87	CODE
LOW	12_gradient_boosting/02_gbm_comparison.py	48	CODE
LOW	12_gradient_boosting/02_gbm_comparison.py	58	CODE
LOW	12_gradient_boosting/10_shap_nlp_sentiment.py	45	CODE
LOW	12_gradient_boosting/03_dl_vs_gbm.py	55	CODE
LOW	12_gradient_boosting/12_case_study_insights.py	64	CODE
LOW	…financial_data_universe/14_point_in_time_validation.py	56	CODE
LOW	02_financial_data_universe/17_complete_pipeline.py	63	CODE
LOW	02_financial_data_universe/20_storage_benchmark_file.py	82	CODE
LOW	16_strategy_simulation/_etf_baseline.py	15	CODE
LOW	…_strategy_simulation/validation/_library_validation.py	3	CODE
LOW	16_strategy_simulation/validation/weights.py	12	CODE
LOW	…trategy_simulation/validation/adapters/lean_adapter.py	3	CODE
LOW	…trategy_simulation/validation/adapters/lean_adapter.py	5	CODE
LOW	…trategy_simulation/validation/adapters/lean_adapter.py	7	CODE
LOW	…trategy_simulation/validation/adapters/ml4t_adapter.py	16	CODE
LOW	…trategy_simulation/validation/adapters/ml4t_adapter.py	25	CODE
LOW	13_dl_time_series/12_case_study_insights.py	72	CODE
LOW	03_market_microstructure/07_itch_stylized_facts.py	70	CODE
LOW	03_market_microstructure/13_algoseek_minute_bars_eda.py	68	CODE
LOW	03_market_microstructure/18_algoseek_jump_detection.py	64	CODE
LOW	03_market_microstructure/10_iex_lob_reconstruction.py	151	CODE
LOW	…rket_microstructure/08_databento_lob_reconstruction.py	55	CODE
LOW	20_strategy_synthesis/02_feature_evaluation.py	47	CODE
LOW	20_strategy_synthesis/08_recommendations.py	49	CODE
LOW	20_strategy_synthesis/08_recommendations.py	49	CODE
LOW	20_strategy_synthesis/04_signal_to_strategy.py	49	CODE
LOW	tests/test_download_coverage.py	13	CODE
LOW	tests/test_backtest_coverage_exclusion.py	19	CODE
LOW	tests/test_allocation_width_asks_for_all.py	29	CODE
LOW	tests/test_artifact_audit_names_its_scope.py	11	CODE
LOW	tests/test_contiguous_run.py	10	CODE
LOW	tests/test_fx_allocation_selection.py	3	CODE
935 more matches not shown…

Cross-Language Confusion130 hits · 779 pts

Severity	File	Line	Snippet	Context
HIGH	05_synthetic_data/00_classical_simulation.py	1898	# only one.** Measured against GBM as the null, jump-diffusion has the	COMMENT
HIGH	12_gradient_boosting/12_case_study_insights.py	1853	# stays null, and so does a family whose selection was scored over a different number	COMMENT
HIGH	12_gradient_boosting/08_shap_analysis.py	588	# NaN is a float value, not a null, so filter it explicitly (a top-k of purely	COMMENT
HIGH	02_financial_data_universe/13_data_quality_framework.py	241	# A genuine null, not NumPy NaN: polars stores np.nan as a float NaN that the null check	COMMENT
HIGH	02_financial_data_universe/07_sp500_options_eda.py	841	# side. A quantile of an empty frame is null, and formatting a null against a percentage	COMMENT
HIGH	13_dl_time_series/12_case_study_insights.py	1095	# grid could not be derived, both comparisons return null on a null, and `filter` drops	COMMENT
HIGH	19_risk_management/03_position_sizing_mae_mfe.py	106	assert close_prices.index.equals(high_prices.index) and close_prices.index.equals(low_prices.index)	CODE
HIGH	19_risk_management/03_position_sizing_mae_mfe.py	107	assert close_prices.columns.equals(high_prices.columns) and close_prices.columns.equals(	CODE
HIGH	19_risk_management/04_factor_exposure.py	158	assert returns.index.equals(ff3_aligned.index) and returns.index.equals(ff5_aligned.index)	CODE
HIGH⚡	tests/test_contiguous_run.py	45	assert got.equals(expected)	CODE
HIGH	tests/test_prediction_timestamp_zone.py	204	assert as_read.equals(as_written)	CODE
HIGH⚡	tests/test_data_quality.py	58	assert out.equals(df)	CODE
HIGH⚡	tests/test_data_quality.py	64	assert out.equals(df)	CODE
HIGH⚡	tests/test_data_quality.py	70	assert out.equals(df)	CODE
HIGH⚡	tests/test_ranking_a_bankrupt_member.py	75	"""That is the other reason a Sharpe is null, and it is still a broken run."""	STRING
HIGH⚡	tests/test_registry_metrics.py	320	undefined, and ``cross_sectional_ic`` returns NaN for that fold.	STRING
HIGH	tests/test_benchmark_grid.py	150	of 100 with factors [8, null, 8] would read as -87.5% then +700%, and the residual	STRING
HIGH	tests/test_backtest_rekey.py	648	is null, `null > tolerance` is null, and polars' `.any()` skips nulls.	STRING
HIGH	tests/test_allocation.py	260	assert prefix.equals(full)	CODE
HIGH	tests/test_deep_learning_adapter.py	390	assert context.expected_keys.equals(expected)	CODE
HIGH⚡	tests/test_candidate_index_retirement.py	118	assert rejoined.equals(index.sort("prediction_hash"))	CODE
HIGH	tests/test_insight_chapter.py	131	assert defaulted.equals(explicit)	CODE
HIGH⚡	…sts/test_candidate_set_ranks_a_bankrupt_member_last.py	89	"""That is the other reason a Sharpe is null, and it is still an ineligible member."""	STRING
HIGH	tests/test_research_contract_execution.py	743	assert stored_trades.equals(original_trades)	CODE
HIGH⚡	tests/test_sp500_options_research.py	1175	assert unchanged.equals(cohorts)	CODE
HIGH⚡	tests/test_backtest_runner_helpers.py	144	assert out.equals(other)	CODE
HIGH	tests/test_backtest_runner_helpers.py	232	assert result.equals(base_weights)	CODE
HIGH	tests/test_full_coverage_selection.py	580	assert restricted.equals(unrestricted)	CODE
HIGH⚡	tests/test_artifact_sidecar_verification.py	79	null, so a caller carrying these into a training spec records what it verified and	STRING
HIGH	tests/test_sp500_options_garch_pit.py	580	"""A block the optimizer never converged on is left null, not filled and not hidden.	STRING
HIGH	tests/test_feature_engineering_coverage.py	5	``rolling_mean`` emits null, ``ml4t.engineer.features.momentum.rsi`` emits NaN.	STRING
HIGH	tests/test_cross_sectional_ic_mean.py	4	Spearman coefficient. ml4t-diagnostic 0.1.1 reports it as NaN and 0.1.2 as null,	STRING
HIGH⚡	tests/test_us_equities_lazy_projection.py	86	assert pushed.equals(materialized)	CODE
HIGH⚡	tests/test_us_equities_lazy_projection.py	98	assert deferred.equals(eager)	CODE
HIGH	tests/test_us_equities_lazy_projection.py	115	assert reduced.equals(projected.filter(pl.col("symbol").is_in(["AAA", "BBB"])))	CODE
HIGH⚡	tests/test_seed_prediction_fixtures.py	294	assert leader_after.equals(leader_before), "the cohort leader's real artifact was overwritten"	CODE
HIGH⚡	tests/test_seed_prediction_fixtures.py	295	assert other_after.equals(other_before), "an ordinary shipped artifact was overwritten"	CODE
HIGH	tests/test_notebook_render.py	171	assert defaulted.equals(explicit)	CODE
HIGH	tests/test_notebook_render.py	172	assert not defaulted.equals(	CODE
HIGH	tests/test_algoseek_convert.py	559	assert df.equals(df.sort("symbol", "timestamp")), "not ordered by (symbol, timestamp)"	CODE
HIGH⚡	tests/test_latent_factors_no_leak.py	805	assert fresh_predictions.equals(cached_predictions)	CODE
HIGH⚡	tests/test_latent_factors_no_leak.py	806	assert fresh_metrics.equals(cached_metrics)	CODE
HIGH	tests/test_common_sample_daily_ic.py	81	"""The second intersection: constant scores make a date undefined, for every model.	STRING
HIGH	23_knowledge_graphs/02_supply_chain_kg_construction.py	1988	.data(graphData.nodes.filter(d => d.group === 1 && d.degree >= 5))	CODE
HIGH	23_knowledge_graphs/02_supply_chain_kg_construction.py	2028	d.fx = null;	CODE
HIGH	23_knowledge_graphs/02_supply_chain_kg_construction.py	2029	d.fy = null;	CODE
HIGH	11_ml_pipeline/05_shap_analysis.py	126	# Replace inf/NaN with null, then drop all nulls	COMMENT
HIGH	11_ml_pipeline/06_conformal_prediction.py	141	# Replace inf/NaN with null, then drop all nulls	COMMENT
HIGH	11_ml_pipeline/03_logistic_classification.py	148	# Replace inf/NaN with null, then drop all nulls	COMMENT
HIGH	11_ml_pipeline/01_ols_inference.py	116	# Replace inf/NaN with null, then drop all nulls	COMMENT
HIGH	11_ml_pipeline/04_nested_cv_hpo.py	130	# A feature that could not be computed arrives as a null, and every row carrying	COMMENT
HIGH	11_ml_pipeline/02_regularization_paths.py	128	# Replace inf/NaN with null, then drop all nulls	COMMENT
HIGH	utils/downloading.py	524	if merged.equals(stored.sort("timestamp")):	CODE
HIGH	utils/modeling.py	2414	null; the `is_finite` filter is belt-and-braces against an environment	STRING
HIGH	08_financial_features/02_microstructure_features.py	376	polars returns a null, not an error, when a correlation or a median has nothing to	STRING
HIGH	07_defining_the_learning_task/06_ic_inference.py	577	"dotted line at the null, which falls on the mode. "	CODE
HIGH	07_defining_the_learning_task/07_multiple_testing.py	392	# P-value histogram (should be uniform under null)	COMMENT
HIGH	07_defining_the_learning_task/05_signal_evaluation.py	1300	# inherits also lives in the null; across blocks the relabelings are independent.	COMMENT
HIGH	…efining_the_learning_task/10_ml4t_library_ecosystem.py	261	# NaN is distinct from null, so guard against both before averaging.	COMMENT
HIGH	…_defining_the_learning_task/08_causal_sanity_checks.py	300	the observed assignment is itself one of the arrangements under the null, and	STRING
70 more matches not shown…

AI Slop Vocabulary243 hits · 704 pts

Severity	File	Line	Snippet	Context
MEDIUM	21_rl_execution_hedging/01_algorithms_comparison.py	151	# it is the single most robust statistical regularity in speculative price	STRING
MEDIUM	05_synthetic_data/00_classical_simulation.py	139	# \| Heston \| Stochastic volatility \| Vol surfaces, leverage effect \|	COMMENT
MEDIUM	05_synthetic_data/00_classical_simulation.py	624	# - $\rho$ = correlation between price and variance shocks (leverage effect)	COMMENT
MEDIUM	09_model_based_features/14_panel_features.py	671	# were nearly identical. That is the trade: a rank is robust because it discards the distances,	COMMENT
MEDIUM	09_model_based_features/14_panel_features.py	917	# the distances between assets, which is what makes them robust and what makes them invent	COMMENT
MEDIUM	09_model_based_features/09_har_rough_volatility.py	620	# residuals. `HC3` relaxes the first assumption with a leverage adjustment. The last two	COMMENT
MEDIUM	22_rag_financial_research/05_10k_rag_assistant.py	965	# \| Evaluation \| Lightweight harness diagnostics \|	COMMENT
MEDIUM	22_rag_financial_research/03_hybrid_retrieval.py	585	# of top-ranked results, making the fusion more robust.	COMMENT
MEDIUM	22_rag_financial_research/04_ragas_evaluation.py	221	# These cover the failure modes the harness is designed to catch:	COMMENT
MEDIUM	22_rag_financial_research/04_ragas_evaluation.py	306	# Each metric targets one of the four harness outputs. Together they	COMMENT
MEDIUM	22_rag_financial_research/04_ragas_evaluation.py	494	# ## 5. The four harness outputs	COMMENT
MEDIUM	22_rag_financial_research/04_ragas_evaluation.py	609	# ## 7. The harness as a chart	COMMENT
MEDIUM	22_rag_financial_research/04_ragas_evaluation.py	684	# the harness doing its job: every gold term was retrieved for every answerable	COMMENT
MEDIUM	12_gradient_boosting/05_cross_library_hpo.py	610	# is the more robust of the two to outliers, which is the reason to expect it to	COMMENT
MEDIUM	12_gradient_boosting/01_ensemble_foundations.py	382	# (§12.2); for robust attributions, see the SHAP analysis in §12.5.	COMMENT
MEDIUM	12_gradient_boosting/08_shap_analysis.py	548	# a robust relationship; declining importance is an early warning of	COMMENT
MEDIUM	12_gradient_boosting/08_shap_analysis.py	710	# explanation is robust to method choice; where they diverge, the	COMMENT
MEDIUM	02_financial_data_universe/13_data_quality_framework.py	370	# has it backwards. MAD is a robust scale estimate, meaning the extreme returns barely move it:	COMMENT
MEDIUM	…nancial_data_universe/08_options_greeks_computation.py	225	# We'll use Brent's method (a robust root-finding algorithm) to solve this.	COMMENT
MEDIUM	02_financial_data_universe/16_provider_comparison.py	359	# For production systems, we need a robust strategy that handles:	COMMENT
MEDIUM	16_strategy_simulation/18_vectorbt_engine_parity.py	99	# comparisons. VectorBT Pro also supplies contract multipliers and futures-style leverage for the	COMMENT
MEDIUM	16_strategy_simulation/02_futures_backtesting.py	531	# Futures require short selling and leverage	COMMENT
MEDIUM	16_strategy_simulation/12_dsr_validation.py	23	# This notebook demonstrates comprehensive backtest validation using the Deflated Sharpe Ratio (DSR) from `ml4t-diag	COMMENT
MEDIUM	16_strategy_simulation/12_dsr_validation.py	1207	# 5. Combine methods: DSR, PBO, and RAS for comprehensive overfitting detection	COMMENT
MEDIUM	13_dl_time_series/04_transformers.py	298	# making the model robust to distribution shift.	COMMENT
MEDIUM	19_risk_management/06_stress_testing.py	470	# The heatmap is the compact committee view: it shows immediately which portfolios are robust	COMMENT
MEDIUM	19_risk_management/07_drift_detection.py	1027	# The `analyze_drift()` function provides a comprehensive analysis combining all methods.	COMMENT
MEDIUM	19_risk_management/04_factor_exposure.py	86	# are made robust to. Daily return residuals carry some, and ignoring it makes every t-statistic	COMMENT
MEDIUM	03_market_microstructure/13_algoseek_minute_bars_eda.py	25	# This notebook provides a comprehensive exploration of the AlgoSeek TAQ minute bar dataset.	COMMENT
MEDIUM	03_market_microstructure/15_itch_lee_ready.py	385	# Run validation across multiple days to get robust statistics.	COMMENT
MEDIUM	03_market_microstructure/17_databento_bar_sampling.py	28	# robust are imbalance-bar parameters across market conditions?	COMMENT
MEDIUM	03_market_microstructure/17_databento_bar_sampling.py	1728	# DataBento's multi-day MBO data enables robust threshold selection that	COMMENT
MEDIUM	03_market_microstructure/18_algoseek_jump_detection.py	167	# is jump-robust: a single large $\|r_i\|$ is multiplied by its small	COMMENT
MEDIUM	03_market_microstructure/18_algoseek_jump_detection.py	244	# intraday volatility regimes and is robust to the jumps it is trying to	COMMENT
MEDIUM	03_market_microstructure/18_algoseek_jump_detection.py	563	# jumps; bipower variation gives a jump-robust continuous-variance	COMMENT
MEDIUM	…ntal_alternative_data/10_institutional_holdings_13f.py	176	# median across a filing's positions is what makes the test robust to one bad row in an otherwise	COMMENT
MEDIUM	20_strategy_synthesis/01_aggregate_synthesis.py	66	# (a missing selection fails loudly); the test harness sets it True so cost/risk	COMMENT
MEDIUM	20_strategy_synthesis/01_aggregate_synthesis.py	71	# backtests, so a partial aggregation cannot be published as a complete one. The test harness	COMMENT
MEDIUM	tests/preset_patches.py	83	# Max configs per family in training menu YAMLs (keep tests fast but comprehensive).	COMMENT
MEDIUM	tests/test_model_registry.py	559	# Resolved through the harness rather than named here: `research_preview=True` below	COMMENT
MEDIUM	tests/test_fixture_registry_identity.py	141	# incomplete. The harness must not be able to reach that path at all.	COMMENT
MEDIUM	tests/overrides.yaml	387	# timeout plus harness overhead, not a runtime. The fine-tuning cell had never completed on	COMMENT
MEDIUM	tests/overrides.yaml	402	# Measured at 20 steps and 100 sentences, all three checkpoints, through this harness with	COMMENT
MEDIUM	tests/overrides.yaml	582	# `n_jobs=-1`, which wins over the harness's OMP_NUM_THREADS cap (tests/pm_helpers.py:113), so	COMMENT
MEDIUM	tests/overrides.yaml	1433	# Nothing asserts the verdict, in this notebook or in the harness, so a fixture rebuild	COMMENT
MEDIUM	tests/overrides.yaml	1630	# The budget is stated rather than left to the harness default so the number has somewhere to	COMMENT
MEDIUM	tests/overrides.yaml	1711	# execution_tier is not canonical, and the harness injects EXECUTION_TIER="preview" for	COMMENT
MEDIUM	tests/overrides.yaml	1827	# harness fallback would fit five names here too. Without it this would register a	COMMENT
MEDIUM	tests/overrides.yaml	1844	# harness fallback would fit five names here too. Without it this would register a	COMMENT
MEDIUM	tests/overrides.yaml	1858	# harness fallback would fit five names here too. Without it this would register a	COMMENT
MEDIUM	tests/overrides.yaml	2411	# execution_tier is not canonical, and the harness injects EXECUTION_TIER="preview" for	COMMENT
MEDIUM	tests/overrides.yaml	2692	# Declared here rather than left to the harness default of max_symbols alone.	COMMENT
MEDIUM	tests/overrides.yaml	2902	# execution_tier is not canonical, and the harness injects EXECUTION_TIER="preview" for any	COMMENT
MEDIUM	tests/overrides.yaml	2995	# the name it declares. The tier comes from the harness: research_preview_parameters sees	COMMENT
MEDIUM	tests/overrides.yaml	2999	# The reduction is stated rather than left to the max_symbols: 5 the harness would	COMMENT
MEDIUM	tests/overrides.yaml	3031	# stated rather than left to the max_symbols the harness would substitute, so the preview	COMMENT
MEDIUM	tests/overrides.yaml	3055	# The harness forces EXECUTION_TIER=preview and WORKSPACE for any notebook declaring both, and	COMMENT
MEDIUM	tests/overrides.yaml	3165	# so the parameter carries the resolver's name rather than the notebook's own. The harness sets	COMMENT
MEDIUM	tests/overrides.yaml	3247	# 07_gbm take the harness default), and the sweep ranks the predictions those stages	COMMENT
MEDIUM	tests/overrides.yaml	3683	# execution_tier is not canonical, and the harness injects EXECUTION_TIER="preview" for any	COMMENT
183 more matches not shown…

Modern AI Meta-Vocabulary165 hits · 494 pts

Severity	File	Line	Snippet	Context
MEDIUM	pyproject.toml	206	# chromadb<1.2, which conflicts with the Ch22 RAG stack.	COMMENT
MEDIUM	05_synthetic_data/06_llm_tabular_great.py	374	# This is a known limitation of autoregressive generation: the model can "hallucinate"	COMMENT
MEDIUM	09_model_based_features/case_study_temporal_summary.py	321	# object is an embedding model and the binding constraint is when it was trained.	COMMENT
MEDIUM	22_rag_financial_research/05_10k_rag_assistant.py	17	# # 10-K Due Diligence Assistant: RAG with Verifiable Citations	COMMENT
MEDIUM	22_rag_financial_research/05_10k_rag_assistant.py	21	# Chapter 22: RAG for Financial Research	COMMENT
MEDIUM	22_rag_financial_research/05_10k_rag_assistant.py	56	# - A CUDA GPU for the embeddings. No API key: the embedding model is local and	COMMENT
MEDIUM	22_rag_financial_research/05_10k_rag_assistant.py	227	# substituting synthetic text. A RAG demo that invents a corpus when the real	COMMENT
MEDIUM	22_rag_financial_research/05_10k_rag_assistant.py	260	# Sets the LLM, embedding model, and chunk strategy. The Chroma collection is	COMMENT
MEDIUM	22_rag_financial_research/05_10k_rag_assistant.py	691	# of this corpus, this embedding model and this distance metric together,	COMMENT
MEDIUM	22_rag_financial_research/05_10k_rag_assistant.py	797	# Numeric questions are where RAG most often fails silently: a fluent model	COMMENT
MEDIUM	22_rag_financial_research/03_hybrid_retrieval.py	21	# Chapter 22: RAG for Financial Research (Section 22.5)	COMMENT
MEDIUM	22_rag_financial_research/03_hybrid_retrieval.py	1288	# RAG pipelines on retrieval, grounding and abstention.	STRING
MEDIUM	22_rag_financial_research/08_rag_security.py	17	# # Security for Financial Document RAG	COMMENT
MEDIUM	22_rag_financial_research/08_rag_security.py	21	# Chapter 22: RAG for Financial Research (Section 22.7)	COMMENT
MEDIUM	22_rag_financial_research/08_rag_security.py	52	# Prerequisites: the RAG pipeline of sections 22.1 to 22.6, and	COMMENT
MEDIUM	22_rag_financial_research/08_rag_security.py	91	# to financial RAG: indirect prompt injection (adversarial instructions hidden	COMMENT
MEDIUM	22_rag_financial_research/README.md	1	# Chapter 22: RAG for Financial Research	COMMENT
MEDIUM	22_rag_financial_research/README.md	49	### 22.7 Diagnosing RAG Pipeline Bottlenecks	COMMENT
MEDIUM	…_financial_research/02_domain_embeddings_comparison.py	17	# # Domain-Specific Embeddings for Financial RAG	COMMENT
MEDIUM	…_financial_research/02_domain_embeddings_comparison.py	21	# Chapter 22: RAG for Financial Research (Section 22.4)	COMMENT
MEDIUM	22_rag_financial_research/01_sec_filing_pipeline.py	19	# Chapter 22: RAG for Financial Research	COMMENT
MEDIUM	22_rag_financial_research/01_sec_filing_pipeline.py	349	# does well here has done well on one topic, so a domain embedding model's	STRING
MEDIUM⚡	22_rag_financial_research/06_esg_rag_vs_finetune.py	385	# ## Approach B: RAG Interface Contract	COMMENT
MEDIUM⚡	22_rag_financial_research/06_esg_rag_vs_finetune.py	387	# A RAG implementation must answer open-ended questions with retrieved evidence,	COMMENT
MEDIUM⚡	22_rag_financial_research/06_esg_rag_vs_finetune.py	392	# Sample ESG questions for RAG	COMMENT
MEDIUM	22_rag_financial_research/06_esg_rag_vs_finetune.py	17	# # ESG Screening and RAG: Implemented Classifier vs Interface Contract	COMMENT
MEDIUM	22_rag_financial_research/06_esg_rag_vs_finetune.py	29	# 2. A RAG assistant. Not built here. What is recorded instead is the	COMMENT
MEDIUM	22_rag_financial_research/06_esg_rag_vs_finetune.py	50	# - State the evidence contract a RAG producer has to meet before its output	COMMENT
MEDIUM	22_rag_financial_research/06_esg_rag_vs_finetune.py	54	# (`data/alternative/news/bloomberg/`). No RAG answers are generated here.	COMMENT
MEDIUM	22_rag_financial_research/06_esg_rag_vs_finetune.py	430	# Run RAG analysis	COMMENT
MEDIUM	22_rag_financial_research/06_esg_rag_vs_finetune.py	442	# ## Comparison: Classification vs RAG	COMMENT
MEDIUM	22_rag_financial_research/06_esg_rag_vs_finetune.py	445	# RAG producer must satisfy.	COMMENT
MEDIUM	22_rag_financial_research/06_esg_rag_vs_finetune.py	495	# \| Question \| Choose Classification \| Choose RAG \|	COMMENT
MEDIUM	22_rag_financial_research/06_esg_rag_vs_finetune.py	517	# RAG contract status	COMMENT
MEDIUM	22_rag_financial_research/06_esg_rag_vs_finetune.py	635	# 5. The RAG side of this notebook is a contract, not a result. No answers	COMMENT
MEDIUM	22_rag_financial_research/06_esg_rag_vs_finetune.py	643	# Book reference: Section 22.8, on choosing between RAG and fine-tuning.	COMMENT
MEDIUM	…_financial_research/07_institutional_holdings_graph.py	21	# Chapter 22: RAG for Financial Research (Section 22.8)	COMMENT
MEDIUM	…_financial_research/07_institutional_holdings_graph.py	1181	# representations, while Chapter 24 uses the same entities in agentic workflows.	COMMENT
MEDIUM⚡	22_rag_financial_research/04_ragas_evaluation.py	17	# # Financial RAG Evaluation Harness	COMMENT
MEDIUM⚡	22_rag_financial_research/04_ragas_evaluation.py	21	# Chapter 22: RAG for Financial Research (Section 22.7)	COMMENT
MEDIUM⚡	22_rag_financial_research/04_ragas_evaluation.py	23	# A RAG pipeline can fail in four unrelated ways - it can retrieve the wrong	COMMENT
MEDIUM	22_rag_financial_research/04_ragas_evaluation.py	53	# Book reference: Section 22.7, on the three RAG failure modes - retrieval,	COMMENT
MEDIUM	22_rag_financial_research/04_ragas_evaluation.py	722	# Next: [`05_10k_rag_assistant`](05_10k_rag_assistant.ipynb) builds a RAG	COMMENT
MEDIUM	12_gradient_boosting/03_dl_vs_gbm.py	363	# * TabPFN: zero-shot: no training, no validation needed.	COMMENT
MEDIUM	12_gradient_boosting/03_dl_vs_gbm.py	551	# TabPFN, zero-shot, no training, no validation	COMMENT
MEDIUM	02_financial_data_universe/13_data_quality_framework.py	518	# `AnomalyManager` orchestrates the three detectors and reports per symbol. It concatenates	COMMENT
MEDIUM	16_strategy_simulation/04_single_asset_ml4t_backtest.py	21	# This notebook introduces the ml4t-backtest event-driven backtesting engine.	COMMENT
MEDIUM	16_strategy_simulation/04_single_asset_ml4t_backtest.py	39	# 2. Run an event-driven backtest with next-open execution and explicit costs.	COMMENT
MEDIUM	16_strategy_simulation/05_stateful_strategies.py	68	# Book Reference: Chapter 16, Section 16.3 - when event-driven simulation is	COMMENT
MEDIUM	16_strategy_simulation/05_stateful_strategies.py	1308	# 2. **The important question is not "vectorized or event-driven?" in the	COMMENT
MEDIUM	16_strategy_simulation/05_stateful_strategies.py	1324	# stateful strategies are most naturally expressed as event-driven simulation.	COMMENT
MEDIUM	16_strategy_simulation/README.md	56	### 16.3 Vectorized and event-driven backtesting	COMMENT
MEDIUM	16_strategy_simulation/03_single_asset_vectorbt.py	33	# Book Reference: Chapter 16, Section 16.3 - vectorized and event-driven engines.	COMMENT
MEDIUM	16_strategy_simulation/03_single_asset_vectorbt.py	694	# strategy implemented with ml4t-backtest's event-driven engine. The main difference	COMMENT
MEDIUM	16_strategy_simulation/03_single_asset_vectorbt.py	715	# - 06_framework_parity: Compare vectorized and event-driven implementations	COMMENT
MEDIUM	16_strategy_simulation/02_futures_backtesting.py	42	# event-driven validation.	COMMENT
MEDIUM⚡	13_dl_time_series/09_foundation_models.py	26	# Two are run here in that zero-shot mode: Chronos (Amazon), which quantises a	COMMENT
MEDIUM⚡	13_dl_time_series/09_foundation_models.py	34	# What the comparison actually asks. The zero-shot models forecast the context	COMMENT
MEDIUM⚡	13_dl_time_series/09_foundation_models.py	37	# the two horizons are not tied together. So this is a probe of whether a zero-shot	COMMENT
MEDIUM⚡	13_dl_time_series/09_foundation_models.py	159	# Design note on horizon and target alignment. The zero-shot baseline	COMMENT
105 more matches not shown…

Cross-File Repetition83 hits · 415 pts

Severity	File	Line	Snippet	Context
HIGH	13_dl_time_series/09_foundation_models.py	0	mean cross-sectional spearman ic over the dates where it is defined. returns the mean and the defined/total date counts.	STRING
HIGH	13_dl_time_series/07_mamba_ssm.py	0	mean cross-sectional spearman ic over the dates where it is defined. returns the mean and the defined/total date counts.	STRING
HIGH	13_dl_time_series/10_uncertainty.py	0	mean cross-sectional spearman ic over the dates where it is defined. returns the mean and the defined/total date counts.	STRING
HIGH	13_dl_time_series/06_tsmixer.py	0	mean cross-sectional spearman ic over the dates where it is defined. returns the mean and the defined/total date counts.	STRING
HIGH	13_dl_time_series/05_tcn.py	0	mean cross-sectional spearman ic over the dates where it is defined. returns the mean and the defined/total date counts.	STRING
HIGH	13_dl_time_series/08_cnn_image_encoding.py	0	mean cross-sectional spearman ic over the dates where it is defined. returns the mean and the defined/total date counts.	STRING
HIGH	13_dl_time_series/04_transformers.py	0	mean cross-sectional spearman ic over the dates where it is defined. returns the mean and the defined/total date counts.	STRING
HIGH	tests/test_allocation_width_asks_for_all.py	0	create table training_runs ( training_hash text primary key, family text, config_name text, label text ); create table p	STRING
HIGH	tests/test_full_coverage_selection.py	0	create table training_runs ( training_hash text primary key, family text, config_name text, label text ); create table p	STRING
HIGH	tests/test_compare_families_solvency.py	0	create table training_runs ( training_hash text primary key, family text, config_name text, label text ); create table p	STRING
HIGH	tests/test_best_reports_concentration.py	0	create table training_runs ( training_hash text primary key, family text, config_name text, label text ); create table p	STRING
HIGH	tests/test_best_reports_the_checkpoint.py	0	create table training_runs ( training_hash text primary key, family text, config_name text, label text ); create table p	STRING
HIGH	tests/test_compare_allocators_ruin.py	0	create table training_runs ( training_hash text primary key, family text, config_name text, label text ); create table p	STRING
HIGH	tests/test_universe_scoped_selection.py	0	create table training_runs ( training_hash text primary key, family text, config_name text, label text ); create table p	STRING
HIGH	tests/test_allocation_width_asks_for_all.py	0	insert into backtest_runs values ( ?, ?, '{"allocation":{"method":"score_weighted"}}', 'signal' )	STRING
HIGH	tests/test_compare_families_solvency.py	0	insert into backtest_runs values ( ?, ?, '{"allocation":{"method":"score_weighted"}}', 'signal' )	STRING
HIGH	tests/test_full_coverage_selection.py	0	insert into backtest_runs values ( ?, ?, '{"allocation":{"method":"score_weighted"}}', 'signal' )	STRING
HIGH	tests/test_latent_factors_identity.py	0	a 64-character hex string here means the source-hashing scheme came back.	STRING
HIGH	tests/test_tabular_dl_identity.py	0	a 64-character hex string here means the source-hashing scheme came back.	STRING
HIGH	tests/test_linear_identity.py	0	a 64-character hex string here means the source-hashing scheme came back.	STRING
HIGH	14_latent_factors/04_ipca.py	0	forecast with an exponentially weighted mean of available factors.	STRING
HIGH	14_latent_factors/06_conditional_autoencoder.py	0	forecast with an exponentially weighted mean of available factors.	STRING
HIGH	14_latent_factors/05_rp_pca.py	0	forecast with an exponentially weighted mean of available factors.	STRING
HIGH	data/equities/market/sp500/build_options_eda.py	0	where the loaders read this dataset from. not ``path(__file__).parent``: the converter writes under ``$ml4t_data_path``,	STRING
HIGH	data/equities/market/sp500/materialize_options.py	0	where the loaders read this dataset from. not ``path(__file__).parent``: the converter writes under ``$ml4t_data_path``,	STRING
HIGH	…a/equities/market/sp500/build_options_straddles_raw.py	0	where the loaders read this dataset from. not ``path(__file__).parent``: the converter writes under ``$ml4t_data_path``,	STRING
HIGH	…studies/crypto_perps_funding/17_holdout_predictions.py	0	select p.prediction_hash, p.training_hash, p.checkpoint_kind, p.checkpoint_value, t.config_name, t.spec_json from predic	STRING
HIGH	case_studies/cme_futures/17_holdout_predictions.py	0	select p.prediction_hash, p.training_hash, p.checkpoint_kind, p.checkpoint_value, t.config_name, t.spec_json from predic	STRING
HIGH	case_studies/utils/strategy_analysis.py	0	select p.prediction_hash, p.training_hash, p.checkpoint_kind, p.checkpoint_value, t.config_name, t.spec_json from predic	STRING
HIGH	case_studies/crypto_perps_funding/06_linear.py	0	render declared parameters for a label: `alpha=1000000.0` reads as `alpha=1e+06`.	STRING
HIGH	case_studies/cme_futures/06_linear.py	0	render declared parameters for a label: `alpha=1000000.0` reads as `alpha=1e+06`.	STRING
HIGH	case_studies/sp500_equity_option_analytics/06_linear.py	0	render declared parameters for a label: `alpha=1000000.0` reads as `alpha=1e+06`.	STRING
HIGH	case_studies/sp500_options/06_linear.py	0	render declared parameters for a label: `alpha=1000000.0` reads as `alpha=1e+06`.	STRING
HIGH	case_studies/fx_pairs/06_linear.py	0	render declared parameters for a label: `alpha=1000000.0` reads as `alpha=1e+06`.	STRING
HIGH	case_studies/us_firm_characteristics/05_linear.py	0	render declared parameters for a label: `alpha=1000000.0` reads as `alpha=1e+06`.	STRING
HIGH	…se_studies/crypto_perps_funding/18_holdout_backtest.py	0	the backtest hashes already registered against one holdout prediction set.	STRING
HIGH	…tudies/nasdaq100_microstructure/19_holdout_backtest.py	0	the backtest hashes already registered against one holdout prediction set.	STRING
HIGH	case_studies/us_equities_panel/21_holdout_backtest.py	0	the backtest hashes already registered against one holdout prediction set.	STRING
HIGH	case_studies/cme_futures/18_holdout_backtest.py	0	the backtest hashes already registered against one holdout prediction set.	STRING
HIGH	case_studies/etfs/19_holdout_backtest.py	0	the backtest hashes already registered against one holdout prediction set.	STRING
HIGH	…studies/us_firm_characteristics/16_holdout_backtest.py	0	the backtest hashes already registered against one holdout prediction set.	STRING
HIGH	…se_studies/crypto_perps_funding/18_holdout_backtest.py	0	select prediction_hash from prediction_sets where split = 'holdout' and training_hash = ? and checkpoint_kind is ? and c	STRING
HIGH	…tudies/nasdaq100_microstructure/19_holdout_backtest.py	0	select prediction_hash from prediction_sets where split = 'holdout' and training_hash = ? and checkpoint_kind is ? and c	STRING
HIGH	case_studies/us_equities_panel/21_holdout_backtest.py	0	select prediction_hash from prediction_sets where split = 'holdout' and training_hash = ? and checkpoint_kind is ? and c	STRING
HIGH	case_studies/cme_futures/18_holdout_backtest.py	0	select prediction_hash from prediction_sets where split = 'holdout' and training_hash = ? and checkpoint_kind is ? and c	STRING
HIGH	case_studies/sp500_options/17_holdout_backtest.py	0	select prediction_hash from prediction_sets where split = 'holdout' and training_hash = ? and checkpoint_kind is ? and c	STRING
HIGH	case_studies/etfs/19_holdout_backtest.py	0	select prediction_hash from prediction_sets where split = 'holdout' and training_hash = ? and checkpoint_kind is ? and c	STRING
HIGH	…studies/us_firm_characteristics/16_holdout_backtest.py	0	select prediction_hash from prediction_sets where split = 'holdout' and training_hash = ? and checkpoint_kind is ? and c	STRING
HIGH	case_studies/fx_pairs/19_strategy_analysis.py	0	select prediction_hash from prediction_sets where split = 'holdout' and training_hash = ? and checkpoint_kind is ? and c	STRING
HIGH	case_studies/fx_pairs/18_holdout_backtest.py	0	select prediction_hash from prediction_sets where split = 'holdout' and training_hash = ? and checkpoint_kind is ? and c	STRING
HIGH	case_studies/nasdaq100_microstructure/05_evaluation.py	0	asymmetric plotly error bars from a pair of interval bounds.	STRING
HIGH	…studies/sp500_equity_option_analytics/05_evaluation.py	0	asymmetric plotly error bars from a pair of interval bounds.	STRING
HIGH	case_studies/fx_pairs/05_evaluation.py	0	asymmetric plotly error bars from a pair of interval bounds.	STRING
HIGH	case_studies/nasdaq100_microstructure/07_gbm.py	0	read the loss function out of a declared configuration name.	STRING
HIGH	case_studies/us_equities_panel/07_gbm.py	0	read the loss function out of a declared configuration name.	STRING
HIGH	case_studies/cme_futures/07_gbm.py	0	read the loss function out of a declared configuration name.	STRING
HIGH	case_studies/sp500_options/07_gbm.py	0	read the loss function out of a declared configuration name.	STRING
HIGH	case_studies/fx_pairs/07_gbm.py	0	read the loss function out of a declared configuration name.	STRING
HIGH	case_studies/etfs/07_gbm.py	0	read the loss function out of a declared configuration name.	STRING
HIGH	case_studies/cme_futures/06_linear.py	0	rank correlation between two labels' orderings, over what both of them charted.	STRING
23 more matches not shown…

Excessive Try-Catch Wrapping319 hits · 299 pts

Severity	File	Line	Snippet	Context
LOW	sitecustomize.py	82	except Exception: # noqa: BLE001 - startup hook: never break the interpreter	CODE
LOW	sitecustomize.py	147	except Exception: # noqa: BLE001 - startup hook: a missing or hung git is not an error here	CODE
LOW	sitecustomize.py	192	except Exception: # noqa: BLE001 - startup hook: never break the interpreter	CODE
LOW	21_rl_execution_hedging/07_backtest_with_impact.py	437	except Exception as exc: # noqa: BLE001 - collected and re-raised below	CODE
LOW	21_rl_execution_hedging/rl_calibration.py	214	except Exception:	CODE
LOW	…_financial_research/02_domain_embeddings_comparison.py	297	except Exception as e:	CODE
LOW	…_financial_research/02_domain_embeddings_comparison.py	336	except Exception as e:	CODE
LOW	12_gradient_boosting/06_optuna_multi_asset.py	127	except Exception as e:	CODE
LOW⚡	12_gradient_boosting/02_gbm_comparison.py	169	except Exception:	CODE
LOW⚡	12_gradient_boosting/02_gbm_comparison.py	176	except Exception:	CODE
LOW⚡	12_gradient_boosting/02_gbm_comparison.py	185	except Exception:	CODE
LOW	12_gradient_boosting/05_cross_library_hpo.py	630	except Exception as e:	CODE
MEDIUM	…nancial_data_universe/21_storage_benchmark_database.py	1074	def questdb_row_count() -> int:	CODE
LOW	…nancial_data_universe/21_storage_benchmark_database.py	281	except Exception as exc:	CODE
LOW	…nancial_data_universe/21_storage_benchmark_database.py	299	except Exception:	CODE
LOW	…nancial_data_universe/21_storage_benchmark_database.py	311	except Exception:	CODE
LOW	…nancial_data_universe/21_storage_benchmark_database.py	331	except Exception:	CODE
LOW	…nancial_data_universe/21_storage_benchmark_database.py	352	except Exception:	CODE
LOW	…nancial_data_universe/21_storage_benchmark_database.py	372	except Exception:	CODE
LOW	…nancial_data_universe/21_storage_benchmark_database.py	426	except Exception as exc:	CODE
LOW	…nancial_data_universe/21_storage_benchmark_database.py	1077	except Exception:	STRING
LOW	…nancial_data_universe/21_storage_benchmark_database.py	1547	except Exception:	STRING
LOW	…nancial_data_universe/21_storage_benchmark_database.py	1745	except Exception as e:	STRING
LOW	…nancial_data_universe/21_storage_benchmark_database.py	1776	except Exception as e:	STRING
LOW	…nancial_data_universe/21_storage_benchmark_database.py	1820	except Exception as e:	STRING
LOW	…_financial_data_universe/22_pandas_polars_benchmark.py	161	except Exception:	CODE
LOW	…_financial_data_universe/22_pandas_polars_benchmark.py	172	except Exception:	CODE
MEDIUM	…nancial_data_universe/08_options_greeks_computation.py	277	print(f"Error: {abs(test_vol - recovered_iv):.2e}")	CODE
LOW	02_financial_data_universe/16_provider_comparison.py	311	except Exception as e:	CODE
LOW	02_financial_data_universe/16_provider_comparison.py	336	except Exception as e:	CODE
LOW	02_financial_data_universe/16_provider_comparison.py	430	except Exception as e:	CODE
LOW	02_financial_data_universe/16_provider_comparison.py	492	except Exception as e:	CODE
LOW	02_financial_data_universe/16_provider_comparison.py	502	except Exception as e:	CODE
LOW	02_financial_data_universe/16_provider_comparison.py	580	except Exception as e:	CODE
LOW	13_dl_time_series/09_foundation_models.py	363	except Exception as e:	CODE
LOW	13_dl_time_series/09_foundation_models.py	402	except Exception as e:	CODE
LOW	13_dl_time_series/09_foundation_models.py	447	except Exception as e:	CODE
LOW	…ket_microstructure/04_itch_order_lifecycle_analysis.py	269	except Exception as e:	CODE
MEDIUM	…ket_microstructure/04_itch_order_lifecycle_analysis.py	270	print(f"Error loading {msg_type}: {e}")	CODE
LOW	03_market_microstructure/01_itch_parser.py	551	except Exception as e:	CODE
LOW⚡	03_market_microstructure/07_itch_stylized_facts.py	242	except Exception:	CODE
LOW⚡	03_market_microstructure/07_itch_stylized_facts.py	245	except Exception:	CODE
LOW⚡	03_market_microstructure/07_itch_stylized_facts.py	250	except Exception:	CODE
LOW⚡	03_market_microstructure/07_itch_stylized_facts.py	265	except Exception:	CODE
LOW⚡	03_market_microstructure/07_itch_stylized_facts.py	274	except Exception:	CODE
LOW⚡	03_market_microstructure/07_itch_stylized_facts.py	277	except Exception:	CODE
LOW⚡	03_market_microstructure/07_itch_stylized_facts.py	282	except Exception:	CODE
LOW	03_market_microstructure/15_itch_lee_ready.py	412	except Exception as e:	CODE
MEDIUM	03_market_microstructure/15_itch_lee_ready.py	413	print(f"Error processing {file.name}: {e}")	CODE
LOW	03_market_microstructure/17_databento_bar_sampling.py	457	except Exception as e:	CODE
MEDIUM	03_market_microstructure/17_databento_bar_sampling.py	458	print(f"Error building {bar_type} bars: {e}")	CODE
LOW	03_market_microstructure/17_databento_bar_sampling.py	646	except Exception:	CODE
LOW	03_market_microstructure/10_iex_lob_reconstruction.py	136	except Exception:	CODE
LOW	03_market_microstructure/10_iex_lob_reconstruction.py	307	except Exception:	CODE
LOW	03_market_microstructure/14_itch_bar_sampling.py	366	except Exception as e:	CODE
LOW	03_market_microstructure/14_itch_bar_sampling.py	721	except Exception as e:	CODE
MEDIUM	03_market_microstructure/05_itch_trading_activity.py	620	print(f"Error loading {msg_type}: {e}")	CODE
LOW	03_market_microstructure/05_itch_trading_activity.py	621	except Exception as e:	CODE
LOW	04_fundamental_alternative_data/05_entity_resolution.py	561	except Exception as exc: # noqa: BLE001 - offline and uncached, or the optional dep is absent	CODE
LOW	…fundamental_alternative_data/02_sec_filing_explorer.py	325	except Exception as exc: # noqa: BLE001 - a source-malformed filing must not stop the scan	CODE
259 more matches not shown…

Empty Error Handlers49 hits · 245 pts

Severity	File	Line	Snippet	Context
HIGH	sitecustomize.py	82	except Exception: # noqa: BLE001 - startup hook: never break the interpreter	CODE
HIGH	sitecustomize.py	192	except Exception: # noqa: BLE001 - startup hook: never break the interpreter	CODE
HIGH	12_gradient_boosting/02_gbm_comparison.py	153	except (FileNotFoundError, subprocess.TimeoutExpired):	CODE
HIGH	12_gradient_boosting/02_gbm_comparison.py	169	except Exception:	CODE
HIGH	12_gradient_boosting/02_gbm_comparison.py	176	except Exception:	CODE
HIGH	12_gradient_boosting/02_gbm_comparison.py	185	except Exception:	CODE
HIGH	03_market_microstructure/07_itch_stylized_facts.py	250	except Exception:	CODE
HIGH	03_market_microstructure/07_itch_stylized_facts.py	265	except Exception:	CODE
HIGH	03_market_microstructure/07_itch_stylized_facts.py	282	except Exception:	CODE
HIGH	tests/test_killed_sequence_run_keeps_its_checkpoints.py	43	except RuntimeError:	CODE
HIGH	tests/pm_helpers.py	1688	except Exception:	CODE
HIGH	23_knowledge_graphs/02_supply_chain_kg_construction.py	385	except (json.JSONDecodeError, KeyError):	CODE
HIGH	utils/reproducibility.py	42	except ImportError:	CODE
HIGH	utils/__init__.py	75	except ImportError:	CODE
HIGH	utils/__init__.py	63	except Exception: # noqa: BLE001	CODE
HIGH	utils/downloading.py	49	except (ImportError, FileNotFoundError):	CODE
HIGH	utils/modeling.py	186	except ImportError:	CODE
HIGH	07_defining_the_learning_task/05_signal_evaluation.py	1782	except ValueError:	CODE
HIGH	scripts/verify_installation.py	43	except ImportError:	CODE
HIGH	scripts/verify_installation.py	550	except Exception:	CODE
HIGH	.github/scripts/notebook_provenance.py	385	except (ValueError, TypeError):	CODE
HIGH	.github/scripts/check_constant_subscripts.py	873	except ValueError:	CODE
HIGH	25_live_trading/13_runtime_safety_showcase.py	619	except asyncio.CancelledError:	CODE
HIGH	25_live_trading/13_runtime_safety_showcase.py	516	except TypeError:	CODE
HIGH	25_live_trading/04_alpaca_paper_trading_demo.py	77	except ImportError:	CODE
HIGH	25_live_trading/09_crypto_funding_deployment_loop.py	501	except ImportError:	CODE
HIGH	25_live_trading/03_ib_paper_trading_demo.py	448	except asyncio.CancelledError:	CODE
HIGH	25_live_trading/05_alpaca_crypto_live_demo.py	74	except ImportError:	CODE
HIGH	data/equities/fundamentals/filings_download.py	259	except Exception:	CODE
HIGH	data/equities/fundamentals/filings_download.py	270	except Exception:	CODE
HIGH	data/equities/market/microstructure/mbo_download.py	213	except Exception:	CODE
HIGH	…ext_feature_engineering/05_financial_ner_finetuning.py	467	except Exception:	CODE
HIGH	case_studies/utils/folds.py	750	except OSError:	CODE
HIGH	case_studies/utils/uncertainty.py	906	except Exception:	CODE
HIGH	case_studies/utils/uncertainty.py	917	except Exception:	CODE
HIGH	case_studies/utils/strategy_analysis.py	2564	except Exception:	CODE
HIGH	case_studies/utils/registry/store.py	544	except sqlite3.DatabaseError:	CODE
HIGH	case_studies/utils/registry/store.py	1376	except ImportError: # pragma: no cover	CODE
HIGH	24_autonomous_agents/agent_research.py	96	except json.JSONDecodeError:	CODE
HIGH	24_autonomous_agents/agent_research.py	109	except json.JSONDecodeError:	CODE
HIGH	24_autonomous_agents/agent_research.py	122	except json.JSONDecodeError:	CODE
HIGH	24_autonomous_agents/agent_research.py	138	except (TypeError, ValueError):	CODE
HIGH	24_autonomous_agents/agent_providers.py	554	except Exception:	CODE
HIGH	24_autonomous_agents/04_research_agent.py	179	except json.JSONDecodeError:	CODE
HIGH	24_autonomous_agents/04_research_agent.py	192	except json.JSONDecodeError:	CODE
HIGH	24_autonomous_agents/04_research_agent.py	205	except json.JSONDecodeError:	CODE
HIGH	24_autonomous_agents/04_research_agent.py	232	except (TypeError, ValueError):	CODE
HIGH	24_autonomous_agents/agent_tools.py	477	except ValueError:	CODE
HIGH	24_autonomous_agents/agent_tools.py	483	except (ValueError, TypeError):	CODE

Deep Nesting253 hits · 210 pts

Severity	File	Line	Context
LOW	…execution_hedging/06_inverse_reinforcement_learning.py	613	CODE
LOW	05_synthetic_data/timegan_metrics.py	555	CODE
LOW	05_synthetic_data/07_dp_gan.py	378	CODE
LOW	…_financial_data_universe/22_pandas_polars_benchmark.py	242	CODE
LOW	16_strategy_simulation/05_stateful_strategies.py	191	CODE
LOW	16_strategy_simulation/validation/weights.py	162	CODE
LOW	…trategy_simulation/validation/adapters/ml4t_adapter.py	299	CODE
LOW	19_risk_management/06_stress_testing.py	1027	CODE
LOW	03_market_microstructure/09_databento_mbo_analysis.py	360	CODE
LOW	03_market_microstructure/09_databento_mbo_analysis.py	394	CODE
LOW	03_market_microstructure/07_itch_stylized_facts.py	220	CODE
LOW	03_market_microstructure/15_itch_lee_ready.py	178	CODE
LOW	03_market_microstructure/15_itch_lee_ready.py	224	CODE
LOW	03_market_microstructure/15_itch_lee_ready.py	270	CODE
LOW	03_market_microstructure/15_itch_lee_ready.py	389	CODE
LOW	03_market_microstructure/17_databento_bar_sampling.py	409	CODE
LOW	03_market_microstructure/limit_orderbook.py	234	CODE
LOW	03_market_microstructure/limit_orderbook.py	484	CODE
LOW	03_market_microstructure/limit_orderbook.py	1016	CODE
LOW	03_market_microstructure/10_iex_lob_reconstruction.py	200	CODE
LOW	…rket_microstructure/08_databento_lob_reconstruction.py	478	CODE
LOW	03_market_microstructure/05_itch_trading_activity.py	593	CODE
LOW	20_strategy_synthesis/08_recommendations.py	294	CODE
LOW	20_strategy_synthesis/08_recommendations.py	444	CODE
LOW	20_strategy_synthesis/05_portfolio_allocation.py	755	CODE
LOW	20_strategy_synthesis/01_aggregate_synthesis.py	1168	CODE
LOW	tests/test_sequence_lookup_reaches_registration.py	66	CODE
LOW	tests/add_missing_parameters_cells.py	44	CODE
LOW	tests/test_fixture_registry_self_consistency.py	127	CODE
LOW	tests/generate_test_microstructure.py	522	CODE
LOW	tests/generate_test_microstructure.py	718	CODE
LOW	tests/generate_test_microstructure.py	942	CODE
LOW	tests/conftest.py	156	CODE
LOW	tests/conftest.py	573	CODE
LOW	tests/sample_registry_for_tests.py	518	CODE
LOW	tests/sample_registry_for_tests.py	714	CODE
LOW	tests/test_notebook_output_hygiene.py	297	CODE
LOW	tests/test_notebook_output_hygiene.py	384	CODE
LOW	tests/test_notebook_output_hygiene.py	478	CODE
LOW	tests/test_model_registry.py	515	CODE
LOW	tests/test_fixture_prediction_panels.py	48	CODE
LOW	tests/test_fixture_prediction_panels.py	67	CODE
LOW	tests/test_ch24_agent_contracts.py	286	CODE
LOW	tests/test_causal_rows_are_resolvable.py	59	CODE
LOW	tests/test_pm_helpers.py	723	CODE
LOW	tests/test_pm_helpers.py	747	CODE
LOW	tests/test_holdout_boundary.py	1030	CODE
LOW	tests/test_leakage_detectors.py	182	CODE
LOW	tests/test_leakage_detectors.py	326	CODE
LOW	tests/test_eoa_universe_roster.py	142	CODE
LOW	tests/pm_helpers.py	236	CODE
LOW	tests/pm_helpers.py	390	CODE
LOW	tests/pm_helpers.py	541	CODE
LOW	tests/pm_helpers.py	624	CODE
LOW	tests/pm_helpers.py	652	CODE
LOW	tests/pm_helpers.py	698	CODE
LOW	tests/pm_helpers.py	802	CODE
LOW	tests/generate_intermediates.py	220	CODE
LOW	tests/test_fold_resolution.py	208	CODE
LOW	tests/test_fold_resolution.py	290	CODE
193 more matches not shown…

Verbosity Indicators39 hits · 77 pts

Severity	File	Line	Snippet	Context
LOW⚡	05_synthetic_data/03_sigcwgan_signatures.py	590	# Step 1: Scale(2, dim=0) - scale first dimension by 2	COMMENT
LOW⚡	05_synthetic_data/03_sigcwgan_signatures.py	593	# Step 2: AddTime - prepend time coordinate	COMMENT
LOW⚡	05_synthetic_data/03_sigcwgan_signatures.py	596	# Step 3: LeadLag - lead-lag transform	COMMENT
LOW⚡	05_synthetic_data/03_sigcwgan_signatures.py	599	# Step 4: VisiTrans("I") - I-visibility transform	COMMENT
LOW⚡	05_synthetic_data/03_sigcwgan_signatures.py	1163	# Step 1: Scale(2, dim=0)	COMMENT
LOW⚡	05_synthetic_data/03_sigcwgan_signatures.py	1166	# Step 2: AddTime	COMMENT
LOW⚡	05_synthetic_data/03_sigcwgan_signatures.py	1169	# Step 3: LeadLag	COMMENT
LOW⚡	05_synthetic_data/03_sigcwgan_signatures.py	1172	# Step 4: VisiTrans("I")	COMMENT
LOW	02_financial_data_universe/19_incremental_updates.py	108	# ### Step 1: Initial Load (Full History)	COMMENT
LOW	02_financial_data_universe/19_incremental_updates.py	124	# ### Step 2: Incremental Update (Only New Data)	COMMENT
LOW⚡	16_strategy_simulation/13_ras_protocol.py	755	# # Step 1: Standardize without demeaning, then compute native estimates	COMMENT
LOW⚡	16_strategy_simulation/13_ras_protocol.py	760	# # Step 2: Apply RAS adjustment	COMMENT
LOW⚡	16_strategy_simulation/13_ras_protocol.py	769	# # Step 3: Annualize only after applying the native-unit bound	COMMENT
LOW⚡	03_market_microstructure/01_itch_parser.py	243	# Step 1: Raw binary from download	COMMENT
LOW⚡	03_market_microstructure/01_itch_parser.py	251	# Step 2: Parsed messages (single uppercase letter = message type)	COMMENT
LOW	14_latent_factors/02_eigenportfolios.py	563	# Step 1: Intra-sector PCA	COMMENT
LOW	14_latent_factors/02_eigenportfolios.py	592	# Step 2: Cross-sector PCA	COMMENT
LOW⚡	…_portfolio_construction/06_hierarchical_risk_parity.py	167	# ### Step 1: Tree Clustering	COMMENT
LOW⚡	…_portfolio_construction/06_hierarchical_risk_parity.py	173	# ### Step 2: Quasi-Diagonalization	COMMENT
LOW⚡	…_portfolio_construction/06_hierarchical_risk_parity.py	177	# ### Step 3: Recursive Bisection	COMMENT
LOW⚡	…_portfolio_construction/06_hierarchical_risk_parity.py	301	# Step 1: Cluster	COMMENT
LOW⚡	…_portfolio_construction/06_hierarchical_risk_parity.py	304	# Step 2: Quasi-diagonalize	COMMENT
LOW⚡	…_portfolio_construction/06_hierarchical_risk_parity.py	307	# Step 3: Recursive bisection	COMMENT
LOW	…_portfolio_construction/06_hierarchical_risk_parity.py	189	# #### Step 1: Hierarchical Clustering	COMMENT
LOW	…_portfolio_construction/06_hierarchical_risk_parity.py	212	# #### Step 2: Leaf Ordering	COMMENT
LOW	…_portfolio_construction/06_hierarchical_risk_parity.py	242	# #### Step 3: Recursive Bisection Allocation	COMMENT
LOW⚡	18_transaction_costs/07_ml4t_volume_participation.py	583	# Step 1: Apply volume limit	COMMENT
LOW⚡	18_transaction_costs/07_ml4t_volume_participation.py	587	# Step 2: Apply market impact	COMMENT
LOW⚡	…efining_the_learning_task/02_preprocessing_pipeline.py	556	# ### Step 1: Remove penny stocks	COMMENT
LOW	…efining_the_learning_task/02_preprocessing_pipeline.py	571	# ### Step 2: Domain filters	COMMENT
LOW	…efining_the_learning_task/02_preprocessing_pipeline.py	599	# ### Step 3: Implausibly large one-day moves	COMMENT
LOW	…efining_the_learning_task/02_preprocessing_pipeline.py	637	# ### Step 4: Spike detection	COMMENT
LOW⚡	15_causal_estimation/02_dowhy_causal_graph.py	250	# Step 1: Base features	COMMENT
LOW⚡	15_causal_estimation/02_dowhy_causal_graph.py	256	# Step 2: Rolling features (backward-looking only)	COMMENT
LOW	15_causal_estimation/02_dowhy_causal_graph.py	267	# Step 3: Treatment (measured at time t) and outcomes (forward-looking)	COMMENT
LOW	15_causal_estimation/02_dowhy_causal_graph.py	280	# Step 4: Binary treatment	COMMENT
LOW⚡	24_autonomous_agents/agent_fixtures.py	541	# Step 2: roll forward within the same event series	COMMENT
LOW	24_autonomous_agents/agent_fixtures.py	526	# Step 1: try the pinned ticker	COMMENT
LOW	24_autonomous_agents/agent_fixtures.py	553	# Step 3: static Fed-policy stand-in	COMMENT

Structural Annotation Overuse39 hits · 76 pts

Severity	File	Line	Snippet	Context
LOW⚡	05_synthetic_data/03_sigcwgan_signatures.py	590	# Step 1: Scale(2, dim=0) - scale first dimension by 2	COMMENT
LOW⚡	05_synthetic_data/03_sigcwgan_signatures.py	593	# Step 2: AddTime - prepend time coordinate	COMMENT
LOW⚡	05_synthetic_data/03_sigcwgan_signatures.py	596	# Step 3: LeadLag - lead-lag transform	COMMENT
LOW⚡	05_synthetic_data/03_sigcwgan_signatures.py	599	# Step 4: VisiTrans("I") - I-visibility transform	COMMENT
LOW⚡	05_synthetic_data/03_sigcwgan_signatures.py	1163	# Step 1: Scale(2, dim=0)	COMMENT
LOW⚡	05_synthetic_data/03_sigcwgan_signatures.py	1166	# Step 2: AddTime	COMMENT
LOW⚡	05_synthetic_data/03_sigcwgan_signatures.py	1169	# Step 3: LeadLag	COMMENT
LOW⚡	05_synthetic_data/03_sigcwgan_signatures.py	1172	# Step 4: VisiTrans("I")	COMMENT
LOW	02_financial_data_universe/19_incremental_updates.py	108	# ### Step 1: Initial Load (Full History)	COMMENT
LOW	02_financial_data_universe/19_incremental_updates.py	124	# ### Step 2: Incremental Update (Only New Data)	COMMENT
LOW⚡	16_strategy_simulation/13_ras_protocol.py	755	# # Step 1: Standardize without demeaning, then compute native estimates	COMMENT
LOW⚡	16_strategy_simulation/13_ras_protocol.py	760	# # Step 2: Apply RAS adjustment	COMMENT
LOW⚡	16_strategy_simulation/13_ras_protocol.py	769	# # Step 3: Annualize only after applying the native-unit bound	COMMENT
LOW⚡	03_market_microstructure/01_itch_parser.py	243	# Step 1: Raw binary from download	COMMENT
LOW⚡	03_market_microstructure/01_itch_parser.py	251	# Step 2: Parsed messages (single uppercase letter = message type)	COMMENT
LOW	14_latent_factors/02_eigenportfolios.py	563	# Step 1: Intra-sector PCA	COMMENT
LOW	14_latent_factors/02_eigenportfolios.py	592	# Step 2: Cross-sector PCA	COMMENT
LOW⚡	…_portfolio_construction/06_hierarchical_risk_parity.py	167	# ### Step 1: Tree Clustering	COMMENT
LOW⚡	…_portfolio_construction/06_hierarchical_risk_parity.py	173	# ### Step 2: Quasi-Diagonalization	COMMENT
LOW⚡	…_portfolio_construction/06_hierarchical_risk_parity.py	177	# ### Step 3: Recursive Bisection	COMMENT
LOW⚡	…_portfolio_construction/06_hierarchical_risk_parity.py	301	# Step 1: Cluster	COMMENT
LOW⚡	…_portfolio_construction/06_hierarchical_risk_parity.py	304	# Step 2: Quasi-diagonalize	COMMENT
LOW⚡	…_portfolio_construction/06_hierarchical_risk_parity.py	307	# Step 3: Recursive bisection	COMMENT
LOW	…_portfolio_construction/06_hierarchical_risk_parity.py	189	# #### Step 1: Hierarchical Clustering	COMMENT
LOW	…_portfolio_construction/06_hierarchical_risk_parity.py	212	# #### Step 2: Leaf Ordering	COMMENT
LOW	…_portfolio_construction/06_hierarchical_risk_parity.py	242	# #### Step 3: Recursive Bisection Allocation	COMMENT
LOW⚡	18_transaction_costs/07_ml4t_volume_participation.py	583	# Step 1: Apply volume limit	COMMENT
LOW⚡	18_transaction_costs/07_ml4t_volume_participation.py	587	# Step 2: Apply market impact	COMMENT
LOW⚡	…efining_the_learning_task/02_preprocessing_pipeline.py	556	# ### Step 1: Remove penny stocks	COMMENT
LOW	…efining_the_learning_task/02_preprocessing_pipeline.py	571	# ### Step 2: Domain filters	COMMENT
LOW	…efining_the_learning_task/02_preprocessing_pipeline.py	599	# ### Step 3: Implausibly large one-day moves	COMMENT
LOW	…efining_the_learning_task/02_preprocessing_pipeline.py	637	# ### Step 4: Spike detection	COMMENT
LOW⚡	15_causal_estimation/02_dowhy_causal_graph.py	250	# Step 1: Base features	COMMENT
LOW⚡	15_causal_estimation/02_dowhy_causal_graph.py	256	# Step 2: Rolling features (backward-looking only)	COMMENT
LOW	15_causal_estimation/02_dowhy_causal_graph.py	267	# Step 3: Treatment (measured at time t) and outcomes (forward-looking)	COMMENT
LOW	15_causal_estimation/02_dowhy_causal_graph.py	280	# Step 4: Binary treatment	COMMENT
LOW⚡	24_autonomous_agents/agent_fixtures.py	541	# Step 2: roll forward within the same event series	COMMENT
LOW	24_autonomous_agents/agent_fixtures.py	526	# Step 1: try the pinned ticker	COMMENT
LOW	24_autonomous_agents/agent_fixtures.py	553	# Step 3: static Fed-policy stand-in	COMMENT

AI Structural Patterns85 hits · 64 pts

Severity	File	Line	Context
LOW	21_rl_execution_hedging/rl_environments.py	66	CODE
LOW	21_rl_execution_hedging/crypto_execution_env.py	42	CODE
LOW	05_synthetic_data/timegan_metrics.py	394	CODE
LOW	05_synthetic_data/timegan_metrics.py	555	CODE
LOW	05_synthetic_data/timegan_metrics.py	703	CODE
LOW	05_synthetic_data/05_diffusion_ts.py	739	CODE
LOW	05_synthetic_data/05_diffusion_ts.py	843	CODE
LOW	05_synthetic_data/03_sigcwgan_signatures.py	602	CODE
LOW	05_synthetic_data/03_sigcwgan_signatures.py	1175	CODE
LOW	…_financial_data_universe/22_pandas_polars_benchmark.py	767	CODE
LOW	…_financial_data_universe/22_pandas_polars_benchmark.py	774	CODE
LOW	…_financial_data_universe/22_pandas_polars_benchmark.py	797	CODE
LOW	…_financial_data_universe/22_pandas_polars_benchmark.py	808	CODE
LOW	…_financial_data_universe/22_pandas_polars_benchmark.py	830	CODE
LOW	…_financial_data_universe/22_pandas_polars_benchmark.py	841	CODE
LOW	…_financial_data_universe/22_pandas_polars_benchmark.py	889	CODE
LOW	…_financial_data_universe/22_pandas_polars_benchmark.py	901	CODE
LOW	…_financial_data_universe/22_pandas_polars_benchmark.py	984	CODE
LOW	…_financial_data_universe/22_pandas_polars_benchmark.py	991	CODE
LOW	…_financial_data_universe/22_pandas_polars_benchmark.py	1028	CODE
LOW	…_financial_data_universe/22_pandas_polars_benchmark.py	1036	CODE
LOW	…_financial_data_universe/22_pandas_polars_benchmark.py	1079	CODE
LOW	…_financial_data_universe/22_pandas_polars_benchmark.py	1097	CODE
LOW	16_strategy_simulation/13_ras_protocol.py	180	CODE
LOW	16_strategy_simulation/05_stateful_strategies.py	319	CODE
LOW	16_strategy_simulation/05_stateful_strategies.py	1030	CODE
LOW	16_strategy_simulation/12_dsr_validation.py	149	CODE
LOW	19_risk_management/10_ml4t_backtest_risk_demo.py	88	CODE
LOW	tests/pm_helpers.py	1521	CODE
LOW	…efining_the_learning_task/02_preprocessing_pipeline.py	212	CODE
LOW	…efining_the_learning_task/02_preprocessing_pipeline.py	335	CODE
LOW	…earning_task/04_maximum_favorable_adverse_excursion.py	237	CODE
LOW	data/futures/loader.py	12	CODE
LOW	data/equities/loader.py	314	CODE
LOW	case_studies/research/execution.py	364	CODE
LOW	case_studies/research/cv.py	272	CODE
LOW	case_studies/crypto_perps_funding/research_workflow.py	538	CODE
LOW	case_studies/config/patchtst/patchtst.py	51	CODE
LOW	case_studies/config/patchtst/_reference/backbone.py	28	CODE
LOW	case_studies/config/patchtst/_reference/backbone.py	193	CODE
LOW	case_studies/config/patchtst/_reference/backbone.py	281	CODE
LOW	case_studies/config/patchtst/_reference/backbone.py	343	CODE
LOW	case_studies/us_equities_panel/03_financial_features.py	369	CODE
LOW	case_studies/us_equities_panel/03_financial_features.py	413	CODE
LOW	case_studies/us_equities_panel/03_financial_features.py	446	CODE
LOW	case_studies/us_equities_panel/03_financial_features.py	487	CODE
LOW	case_studies/us_equities_panel/03_financial_features.py	544	CODE
LOW	case_studies/us_equities_panel/03_financial_features.py	566	CODE
LOW	case_studies/us_equities_panel/03_financial_features.py	633	CODE
LOW	case_studies/utils/deep_learning.py	2297	CODE
LOW	case_studies/utils/signals.py	324	CODE
LOW	case_studies/utils/sequence_dataset.py	912	CODE
LOW	case_studies/utils/coverage.py	997	CODE
LOW	case_studies/utils/notebook_render.py	370	CODE
LOW	case_studies/utils/tabular_dl.py	2345	CODE
LOW	case_studies/utils/tabular_dl.py	2484	CODE
LOW	case_studies/utils/backtest_presets.py	524	CODE
LOW	case_studies/utils/slot_strategy.py	228	CODE
LOW	case_studies/utils/causal.py	444	CODE
LOW	case_studies/utils/causal.py	951	CODE
25 more matches not shown…

Docstring Block Structure11 hits · 55 pts

Severity	File	Line	Snippet	Context
HIGH	tests/pm_helpers.py	1534	Execute a notebook via Papermill with parameter injection. This is the core test helper. It: 1. Syncs .py -> .i	STRING
HIGH	tests/create_test_data.py	1561	Copy production's files for one downloaded source, verbatim, and check the minimum. Args: source: Productio	STRING
HIGH	utils/paths.py	168	Get the directory path for a chapter. Args: chapter: Chapter number (1-27) Returns: Absolute p	STRING
HIGH	utils/paths.py	204	Get output directory for cross-chapter data flow. This is the primary function for case study data that flows betwe	STRING
HIGH	utils/paths.py	303	Get the case study directory for a strategy. Case studies are centralized under CASE_STUDIES_DIR (default: repo_roo	STRING
HIGH	utils/data_quality.py	622	Run all data quality checks before modeling. This is the gate between data preparation (labels + features) and	STRING
HIGH	utils/storage_benchmarks.py	331	Block until an asynchronously-ingesting engine reports `expected_rows`. QuestDB (ILP + WAL) and InfluxDB acknowledg	STRING
HIGH	data/futures/loader.py	84	List CME product codes available in the local data store. Args: frequency: ``"hourly"`` (default) lists pro	STRING
HIGH	data/futures/loader.py	315	Load CFTC Commitment of Traders (COT) data. Reads per-product parquets written by ``data/futures/positioning/cot_do	STRING
HIGH	data/fx/loader.py	13	List currency pairs available in the local data store. Args: frequency: Which parquet to probe (``"daily"``	STRING
HIGH	data/etfs/loader.py	93	Load the traded close and share count, unadjusted for splits and distributions. :func:`load_etfs` returns an adjust	STRING

Modern Structural Boilerplate50 hits · 48 pts

Severity	File	Line	Snippet	Context
LOW	16_strategy_simulation/05_stateful_strategies.py	689	def _update_pair_position(self, zscore: float, price_a: float, price_b: float, broker) -> None:	CODE
LOW	16_strategy_simulation/05_stateful_strategies.py	977	def _update_circuit_state(self, equity: float) -> None:	CODE
LOW	tests/test_causal_adapter.py	171	def _set_holdout_start(setup_path, value: str) -> None:	CODE
LOW	tests/test_weekly_issues.py	91	def set_state(self, number: int, state: str, reason: str \| None = None) -> None:	CODE
LOW⚡	utils/style.py	1053	__all__ = [	CODE
LOW	utils/paths.py	444	__all__ = [	CODE
LOW	utils/reproducibility.py	15	def set_global_seeds(seed: int = 42) -> None:	CODE
LOW	utils/__init__.py	78	__all__ = [	CODE
LOW	utils/artifact_specs.py	217	__all__ = [	CODE
LOW	17_portfolio_construction/deepm/__init__.py	22	__all__ = [	CODE
LOW	17_portfolio_construction/deepm/utils.py	13	def set_seed(seed: int) -> None:	CODE
LOW	scripts/download_artifacts.py	283	def _set_tree_writable(root: Path, *, writable: bool) -> None:	CODE
LOW	.github/scripts/weekly_issues.py	279	def set_state(self, number: int, state: str, reason: str \| None = None) -> None: ...	CODE
LOW	.github/scripts/weekly_issues.py	356	def set_state(self, number: int, state: str, reason: str \| None = None) -> None:	CODE
LOW	25_live_trading/10_safety_risk_demo.py	86	logger = logging.getLogger(__name__)	CODE
LOW	25_live_trading/13_runtime_safety_showcase.py	86	logger = logging.getLogger(__name__)	CODE
LOW	25_live_trading/04_alpaca_paper_trading_demo.py	448	def update_market(self, timestamp: datetime, prices: dict[str, float]) -> None:	CODE
LOW	25_live_trading/01_unified_framework_demo.py	408	def update_price(self, asset: str, price: float, timestamp: datetime) -> None:	CODE
LOW	25_live_trading/03_ib_paper_trading_demo.py	85	logger = logging.getLogger(__name__)	CODE
LOW	25_live_trading/05_alpaca_crypto_live_demo.py	413	def update_market(self, timestamp: datetime, prices: dict[str, float]) -> None:	CODE
LOW	data/__init__.py	126	__all__ = [	CODE
LOW	case_studies/research/__init__.py	91	__all__ = [	CODE
LOW	case_studies/config/patchtst/_reference/backbone.py	6	__all__ = ["PatchTST_backbone"]	CODE
LOW	case_studies/config/patchtst/_reference/__init__.py	21	__all__ = ["PatchTST_backbone"]	CODE
LOW	case_studies/config/patchtst/_reference/layers.py	3	__all__ = [	CODE
LOW	case_studies/utils/feature_engineering.py	36	__all__ = [	CODE
LOW	case_studies/utils/temporal.py	36	__all__ = [	CODE
LOW	case_studies/utils/folds.py	47	__all__ = [	CODE
LOW	case_studies/utils/feasibility.py	14	__all__ = ["cross_sectional_persistence", "exceedance_curve", "fold_timeline", "panel_acf"]	CODE
LOW	case_studies/utils/coverage.py	50	__all__ = [	CODE
LOW	case_studies/utils/notebook_render.py	829	__all__ = [	STRING
LOW	case_studies/utils/runtime.py	34	__all__ = [	CODE
LOW	case_studies/utils/uncertainty.py	1579	__all__ = [	CODE
LOW	case_studies/utils/artifact_quality.py	31	__all__ = [	CODE
LOW	case_studies/utils/derived_params.py	18	__all__ = ["DERIVED_PARAM_SIGNIFICANT_DIGITS", "quantize_derived"]	CODE
LOW	case_studies/utils/persistent_panel.py	64	__all__ = [	CODE
LOW	case_studies/utils/latent_factors/macro_context.py	78	__all__ = ["MACRO_CONTEXT_ALIGNMENT", "load_configured_macro_context"]	CODE
LOW	case_studies/utils/latent_factors/__init__.py	41	__all__ = [	CODE
LOW	case_studies/utils/latent_factors/common.py	167	__all__ = [	CODE
LOW	case_studies/utils/latent_factors/case_study.py	24	logger = logging.getLogger(__name__)	CODE
LOW	case_studies/utils/latent_factors/panel.py	301	__all__ = [	CODE
LOW	case_studies/utils/registry/registration.py	44	logger = logging.getLogger(__name__)	CODE
LOW	case_studies/utils/registry/specs.py	16	logger = logging.getLogger(__name__)	CODE
LOW⚡	case_studies/utils/registry/store.py	25	logger = logging.getLogger(__name__)	CODE
LOW	case_studies/utils/registry/metrics.py	9	logger = logging.getLogger(__name__)	CODE
LOW	case_studies/utils/registry/__init__.py	146	__all__ = [	CODE
LOW	case_studies/utils/registry/completeness.py	653	__all__ = [	CODE
LOW	case_studies/utils/registry/backtest_rekey.py	48	__all__ = [	CODE
LOW	case_studies/utils/registry/queries.py	34	logger = logging.getLogger(__name__)	CODE
LOW	case_studies/sp500_options/_underlying_returns.py	15	__all__ = ["reconcile_underlying_log_returns", "validate_reconciled_returns"]	CODE

Redundant / Tautological Comments39 hits · 48 pts

Severity	File	Line	Snippet	Context
LOW	05_synthetic_data/04_gtgan_irregular.py	1609	# Check if interpolated values are bounded by adjacent real values (soft check)	COMMENT
LOW	05_synthetic_data/04_gtgan_irregular.py	187	RETRAIN = False # Set True to retrain even if checkpoint exists	CODE
LOW	05_synthetic_data/timegan_metrics.py	232	# Check if data appears to be normalized to [0, 1]	COMMENT
LOW	05_synthetic_data/05_diffusion_ts.py	189	RETRAIN = False # Set True to retrain even if checkpoint exists	CODE
LOW	05_synthetic_data/06_llm_tabular_great.py	135	RETRAIN = False # Set True to retrain even if checkpoint exists	CODE
LOW	05_synthetic_data/03_sigcwgan_signatures.py	1341	# Check if checkpoint exists and RETRAIN is False	COMMENT
LOW	05_synthetic_data/03_sigcwgan_signatures.py	189	RETRAIN = False # Set True to force retraining even if checkpoint exists	CODE
LOW	05_synthetic_data/07_dp_gan.py	146	RETRAIN = False # Set True to retrain even if checkpoint exists	CODE
LOW	05_synthetic_data/02_tailgan_tail_risk.py	88	RETRAIN = False # Set True to force retraining even if checkpoint exists	CODE
LOW	05_synthetic_data/02_tailgan_tail_risk.py	561	# Check if constraint violated: W*v >= e	COMMENT
LOW	…nancial_data_universe/21_storage_benchmark_database.py	1713	# Check if process started successfully	STRING
LOW	…nancial_data_universe/08_options_greeks_computation.py	250	# Check if solution exists within bounds	COMMENT
LOW	19_risk_management/07_drift_detection.py	1646	# Check if retraining needed	COMMENT
LOW	…ket_microstructure/04_itch_order_lifecycle_analysis.py	128	# Check if we have message data to analyze	COMMENT
LOW	03_market_microstructure/14_itch_bar_sampling.py	601	# Check if we have actual buy/sell variance	COMMENT
LOW	03_market_microstructure/05_itch_trading_activity.py	109	# Check if we have message data to analyze	COMMENT
LOW	03_market_microstructure/05_itch_trading_activity.py	648	# Print message type breakdown	COMMENT
LOW	tests/test_model_registry.py	587	# Check if this notebook is expected to register (match on suffix)	COMMENT
LOW	23_knowledge_graphs/02_supply_chain_kg_construction.py	83	RERUN_EXTRACTION = False # Set True to force LLM re-extraction; False loads cached triples	CODE
LOW	utils/data_quality.py	672	# Print results	COMMENT
LOW⚡	utils/storage_benchmarks.py	739	# Assign quotes to symbols (weighted: some symbols more active)	COMMENT
LOW⚡	utils/storage_benchmarks.py	804	# Assign trades to symbols (same distribution as quotes)	COMMENT
LOW	08_financial_features/07_event_studies.py	186	# Check if new high	COMMENT
LOW	15_causal_estimation/05_momentum_causal_trading.py	455	# Set it to `False` while iterating on the strategy code below, where the causal estimates	COMMENT
LOW	data/futures/market/download.py	234	# Check if year is complete	COMMENT
LOW	data/equities/loader.py	1370	# Check if raw files exist but aren't parsed	COMMENT
LOW	data/equities/market/us_equities/download.py	90	# Check if already exists	COMMENT
LOW	10_text_feature_engineering/02_asset_embeddings.py	697	# Display results	COMMENT
LOW	case_studies/nasdaq100_microstructure/14_backtest.py	109	FORCE_REBACKTEST = False # Set True to re-backtest even if a complete backtest_hash exists	CODE
LOW	case_studies/utils/model_analysis.py	204	# Check if fold_metrics table exists	STRING
LOW	case_studies/utils/registry/store.py	836	# Check if backtest_runs table exists at all	COMMENT
LOW	case_studies/utils/registry/store.py	1160	# Check if this table still has tall format	COMMENT
LOW	case_studies/utils/registry/queries.py	1640	# Check if stage column exists (the CREATE TABLE IF NOT EXISTS	STRING
LOW	…e_studies/sp500_equity_option_analytics/14_backtest.py	102	FORCE_REBACKTEST = False # Set True to re-backtest even if a complete backtest_hash exists	CODE
LOW	case_studies/us_firm_characteristics/11_backtest.py	98	FORCE_REBACKTEST = False # Set True to re-backtest even if a complete backtest_hash exists	CODE
LOW	26_mlops_governance/01_drift_monitoring.py	49	# Set it to a label and that label is required: the notebook stops rather than monitoring a	COMMENT
LOW	…gents/operator_artifacts/run_etfs_20260504T223150.json	5337	"stdout_tail": "24:# prediction into a backtest across the full prediction \u00d7 entry-scheme grid and\n28:# **	CODE
LOW	…gents/operator_artifacts/run_etfs_20260504T223150.json	5402	"command": "cd ~/ml4t/code && uv run python -c \"\nimport polars as pl\nfrom pathlib import Path\n\nSANDBOX = Pa	CODE
LOW⚡	…facts/run_us_firm_characteristics_20260504T225521.json	2553	"content": "# ---\n# jupyter:\n# jupytext:\n# text_representation:\n# extension: .py\n# format	CODE

Self-Referential Comments13 hits · 36 pts

Severity	File	Line	Snippet	Context
MEDIUM	21_rl_execution_hedging/07_backtest_with_impact.py	335	# ## Define the Momentum Strategy	COMMENT
MEDIUM	05_synthetic_data/07_dp_gan.py	399	# Create a clean discriminator copy for generator training (no DP hooks)	COMMENT
MEDIUM	22_rag_financial_research/05_10k_rag_assistant.py	362	# ### Create the Query Interface	COMMENT
MEDIUM	tests/smoke.yaml	8	# This file is the DECLARATION. The machinery it drives already exists and is not duplicated here:	COMMENT
MEDIUM	23_knowledge_graphs/02_supply_chain_kg_construction.py	214	# Define the knowledge graph schema. Each extracted relationship is a	COMMENT
MEDIUM	23_knowledge_graphs/02_supply_chain_kg_construction.py	1619	# Create the pyvis network with supplier risk tooltips and compact company	COMMENT
MEDIUM	23_knowledge_graphs/08_8k_event_extraction.py	335	# Define the event quadruple structure.	COMMENT
MEDIUM	17_portfolio_construction/08_library_comparison.py	473	# Create the portfolio object and pass the already-defined sample moments explicitly.	COMMENT
MEDIUM	…transaction_costs/12_commission_slippage_comparison.py	638	# ### Define the Equity-Compatible Commission Variants	COMMENT
MEDIUM	07_defining_the_learning_task/03_label_methods.py	596	# This method is from De Prado's Advances in Financial Machine Learning.	COMMENT
MEDIUM	15_causal_estimation/09_adia_causal_benchmark.py	155	# Define the 8 causal categories	COMMENT
MEDIUM	15_causal_estimation/05_momentum_causal_trading.py	191	# ### Defining the Market Volatility Regime	COMMENT
MEDIUM	case_studies/utils/model_analysis.py	837	# Create a key for merge	STRING

Magic Placeholder Names4 hits · 30 pts

Severity	File	Line	Snippet	Context
HIGH	data/futures/market/dataset_card.py	93	# DATABENTO_API_KEY=db-your-api-key-here	COMMENT
HIGH	data/fx/market/dataset_card.py	81	# OANDA_API_KEY=your-api-key-here	COMMENT
HIGH	data/equities/market/microstructure/dataset_card.py	86	# DATABENTO_API_KEY=db-your-api-key-here	COMMENT
HIGH	data/equities/market/us_equities/dataset_card.py	80	# QUANDL_API_KEY=your-api-key-here	COMMENT

Synthetic Comment Markers2 hits · 12 pts

Severity	File	Line	Snippet	Context
HIGH	09_model_based_features/07_arima_features.py	263	# returns as many values as you asked for rather than one per row. The helper passes	COMMENT
HIGH	06_strategy_definition/exploration.md	35	├── figures/ # Book figures (AI-generated + notebook)	CODE

Example Usage Blocks6 hits · 9 pts

Severity	File	Line	Snippet	Context
LOW	tests/generate_test_data.sh	31	# Usage:	COMMENT
LOW	data/crypto/market/config.yaml	11	# Usage:	COMMENT
LOW	data/prediction_markets/config.yaml	8	# Usage:	COMMENT
LOW	data/fx/market/config.yaml	10	# Usage:	COMMENT
LOW	data/macro/config.yaml	10	# Usage:	COMMENT
LOW	data/etfs/market/config.yaml	17	# Usage:	COMMENT

Fake / Example Data5 hits · 5 pts

Severity	File	Line	Snippet	Context
LOW	05_synthetic_data/07_dp_gan.py	441	fake_data = generator(z)	CODE
LOW	05_synthetic_data/07_dp_gan.py	445	d_fake = discriminator_private(fake_data)	CODE
LOW	05_synthetic_data/07_dp_gan.py	469	fake_data = generator(z)	CODE
LOW	05_synthetic_data/07_dp_gan.py	472	d_fake = discriminator_for_g(fake_data)	CODE
LOW⚡	tests/test_pm_helpers.py	147	monkeypatch.setenv("ML4T_FAKE_KEY", "Jane Doe jane@example.org")	CODE

Overly Generic Function Names5 hits · 4 pts

Severity	File	Line	Snippet	Context
LOW⚡	tests/test_notebook_output_hygiene.py	689	py_source = "# %% [markdown]\n# Some prose.\n\n\ndef helper():\n return 1\n"	CODE
LOW⚡	tests/test_notebook_output_hygiene.py	694	"source": ["Some prose.\n", "\n", "def helper():\n", " return 1\n"],	CODE
LOW	tests/test_notebook_output_hygiene.py	708	{"cell_type": "code", "source": ["def helper():\n", " return 1\n"], "metadata": {}},	CODE
LOW	tests/test_import_resolution.py	246	(pkg / "__init__.py").write_text("CONSTANT = 3\n\n\ndef helper():\n pass\n")	CODE
LOW	tests/test_import_resolution.py	471	"def helper():\n from pathlib import Path\n", # binds inside the function only	CODE

Slop Phrases1 hit · 2 pts

Severity	File	Line	Snippet	Context
MEDIUM	…/01_react_reasoning_20260615T191047Z_04eb6e7c603d.json	217	"snippet": "# US - FedWatch Probability of Rate Hike/Cut \| US Fed \| Collection \| MacroMicro. #### Notifica	CODE

Maintenance and Governance Review

Risk Review · v3

What These Metrics Mean

Pattern-Signal History

Severity Breakdown

Directory Score Breakdown

Pattern Findings