Python Core Workflow

Workflow Objective

Run the core research loop in Python while preserving OpenQuant’s leakage and diagnostics standards.

Stage 1: Ingestion and quality controls

Namespace: openquant.data

Primary APIs:

• load_ohlcv
• clean_ohlcv
• align_calendar
• data_quality_report

Stage 2: Event-driven bars

Namespace: openquant.bars

Primary APIs:

• build_time_bars
• build_tick_bars
• build_volume_bars
• build_dollar_bars

Stage 3: Feature diagnostics

Namespace: openquant.feature_diagnostics

Primary APIs:

• mdi_importance
• mda_importance
• sfi_importance
• orthogonalize_features_pca
• substitution_effect_report

Stage 4: Pipeline execution and reporting

Namespace: openquant.pipeline

Primary APIs:

• run_mid_frequency_pipeline
• run_mid_frequency_pipeline_frames
• summarize_pipeline

Quality Criteria

• Feature diagnostics use leakage-safe validation defaults.
• Report artifacts include trial and split metadata.
• Results are reproducible from declared data and code revisions.