Skip to content

CODE_MANIFEST.md

Latest commit

 

History

History
45 lines (29 loc) · 3.3 KB

CODE_MANIFEST.md

File metadata and controls

45 lines (29 loc) · 3.3 KB
Raw

Code manifest

Dataset overview

  • analysis/00_dataset_overview/code/create_dataset_overview_dictionary.py: creates dataset-level summaries and variable dictionaries.

Finished-product issue definition

  • analysis/01_finished_product_issue/code/run_d2_description.R: describes finished-product quality attributes and monthly trends.
  • analysis/01_finished_product_issue/code/run_d2_issue_group_single_plots.R: creates separate plots comparing predefined finished-product issue groups.

Layer-wise description and association

  • analysis/02_tablet_mes_description/code/run_d3_description.R: describes tablet MES production records.
  • analysis/03_tablet_mes_association/code/run_d3_mes_disintegration_association.R: screens tablet MES process variables against finished-product disintegration.
  • analysis/04_extract_powder_description/code/run_d4_description.R: describes Jianwei Xiaoshi extract-powder quality data.
  • analysis/05_extract_powder_association/code/run_d4_finished_disintegration_association.R: links extract-powder quality data to finished-product disintegration.
  • analysis/06_chenpi_description_and_association/code/run_d6_finished_disintegration_association.R: links Chenpi quality records to downstream finished-product disintegration.
  • analysis/06_chenpi_description_and_association/code/run_d6_origin_comparison.R: compares Chenpi quality attributes by source/origin.
  • analysis/07_yam_powder_description/code/run_d7_description.R: describes Chinese yam powder MES production records.
  • analysis/08_yam_powder_association/code/run_d7_finished_disintegration_association.R: links Chinese yam powder MES variables to downstream finished-product disintegration.

Modeling

  • analysis/10_joint_modeling/code/run_joint_disintegration_modeling.R: builds hierarchical diagnostic models.
  • analysis/10_joint_modeling/code/audit_time_and_origin_design_inputs.R: audits temporal/origin variables for later sensitivity analysis.
  • analysis/10_joint_modeling/code/audit_confounding_batch_structure.R: audits upstream-downstream batch multiplicity and confounding-aware analysis feasibility.

Temporal and confounding-aware analysis

  • analysis/11_temporal_analysis/code/run_finished_disintegration_time_series.R: analyzes monthly disintegration patterns and abnormal temporal windows.
  • analysis/12_confounding_aware_analysis/code/run_confounding_aware_batch_decomposition.R: decomposes batch-level evidence under temporal-window exposure.
  • analysis/12_confounding_aware_analysis/code/run_confounding_aware_robust_path_analysis.R: performs confounding-aware robustness and path-attenuation analysis.

Graph evidence scoring

  • analysis/13_graph_evidence_scoring/code/run_batch_evidence_graph_scoring.R: computes graph-based evidence scores and upstream risk-priority rankings.

Public framework asset

  • assets/framework_design.svg: public schematic of the analysis framework. It is included for orientation only and does not contain confidential data.

Public-release boundary

This repository intentionally excludes raw data, standardized analysis-ready data, generated result workbooks, draft documents, writing utilities, and confidential batch-level outputs. The code documents the analysis workflow and public framework design. Users need authorized local datasets to reproduce the full analysis.