# nandita005/decodex_final architecture

> Architecture diagram of the GitHub repository nandita005/decodex_final, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-08). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/nandita005/decodex_final
- Repository: https://github.com/nandita005/decodex_final

## Overview

This is a regulatory-surveillance analysis pipeline: raw exchange-style orders, trades, and suspicious lists become cleaned scrip/date batches, client networks and microstructure signals, then ranked and explainable analyst-review artifacts. The main runner owns the documented stage flow; sampled code confirms key calls, while README statements support several unsampled subsystems. The graph preserves those documented capabilities; links without direct sampled call evidence are omitted or marked unverified.

Main entry points: [`run_pipeline.py`](https://github.com/nandita005/decodex_final/blob/master/run_pipeline.py) (Pipeline runner), [`src/reasoning_engine/signal_fusion.py`](https://github.com/nandita005/decodex_final/blob/master/src/reasoning_engine/signal_fusion.py) (Signal fusion), [`outputs`](https://github.com/nandita005/decodex_final/tree/master/outputs) (Reports and graphs), [`src/preprocessing/load_data.py`](https://github.com/nandita005/decodex_final/blob/master/src/preprocessing/load_data.py) (Load and clean)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_ingest["Data preparation"]
  node_loader["Load and clean<br/>[load_data.py]"]
  node_trade_clean["Trade cleaning<br/>[clean_trades.py]"]
  node_orders_clean["Order cleaning<br/>[clean_orders.py]"]
  node_batch["Scrip/date batches"]
end

subgraph group_signals["Detection signals"]
  node_trade_graph["Trade graph"]
  node_cycle["Cycle detection<br/>[cycle_detection.py]"]
  node_centrality["Centrality and reciprocity"]
  node_network_extensions["Network analysis"]
  node_micro["Microstructure features<br/>[price_volume.py]"]
end

subgraph group_assessment["Risk assessment"]
  node_fusion["Signal fusion<br/>[signal_fusion.py]"]
  node_config["Risk configuration<br/>[config.py]"]
  node_risk["Client and member risk<br/>[client_risk.py]"]
  node_classifier["Mechanism labels"]
end

subgraph group_review["Evidence and review"]
  node_evidence["Evidence records"]
  node_integrate["Evidence integration"]
  node_composite["Composite scoring"]
  node_escalation["Escalation selection"]
  node_explanations["Review explanations"]
end

node_analyst(("Analyst"))
node_raw[("Raw market files")]
node_runner["Pipeline runner<br/>[run_pipeline.py]"]
node_outputs[("Reports and graphs")]

node_raw -->|"loads files"| node_loader
node_runner -->|"loads data"| node_loader
node_loader -->|"cleans trades"| node_trade_clean
node_loader -->|"cleans orders"| node_orders_clean
node_runner -->|"builds batches"| node_batch
node_runner -->|"builds graph"| node_trade_graph
node_trade_graph -->|"normalizes IDs"| node_trade_clean
node_runner -->|"detects cycles"| node_cycle
node_runner -->|"computes metrics"| node_centrality
node_runner -->|"computes features"| node_micro
node_runner -->|"fuses signals"| node_fusion
node_fusion -.->|"uses metrics"| node_centrality
node_fusion -.->|"uses features"| node_micro
node_fusion -.->|"uses cycle nodes"| node_cycle
node_runner -->|"scores clients"| node_risk
node_risk -->|"reads components"| node_fusion
node_risk -->|"validates weights"| node_config
node_runner -->|"classifies clients"| node_classifier
node_runner -->|"generates evidence"| node_evidence
node_evidence -->|"uses signal names"| node_fusion
node_runner -->|"runs stage three"| node_integrate
node_runner -->|"computes final risk"| node_composite
node_runner -->|"selects escalation"| node_escalation
node_runner -->|"writes explanations"| node_explanations
node_integrate -->|"writes combined records"| node_outputs
node_composite -->|"writes scores"| node_outputs
node_explanations -->|"writes report"| node_outputs
node_runner -->|"writes artifacts"| node_outputs
node_trade_graph -.->|"renders graphs"| node_outputs
node_network_extensions -.->|"supports review"| node_analyst
node_outputs -->|"provides review artifacts"| node_analyst

click node_raw "https://github.com/nandita005/decodex_final/tree/master/data/raw"
click node_runner "https://github.com/nandita005/decodex_final/blob/master/run_pipeline.py"
click node_loader "https://github.com/nandita005/decodex_final/blob/master/src/preprocessing/load_data.py"
click node_trade_clean "https://github.com/nandita005/decodex_final/blob/master/src/preprocessing/clean_trades.py"
click node_orders_clean "https://github.com/nandita005/decodex_final/blob/master/src/preprocessing/clean_orders.py"
click node_batch "https://github.com/nandita005/decodex_final/blob/master/src/preprocessing/build_scrip_date_batches.py"
click node_trade_graph "https://github.com/nandita005/decodex_final/blob/master/src/network_analysis/build_trade_graph.py"
click node_cycle "https://github.com/nandita005/decodex_final/blob/master/src/network_analysis/cycle_detection.py"
click node_centrality "https://github.com/nandita005/decodex_final/blob/master/src/network_analysis/centrality_metrics.py"
click node_network_extensions "https://github.com/nandita005/decodex_final/tree/master/src/network_analysis"
click node_micro "https://github.com/nandita005/decodex_final/blob/master/src/microstructure/price_volume.py"
click node_fusion "https://github.com/nandita005/decodex_final/blob/master/src/reasoning_engine/signal_fusion.py"
click node_config "https://github.com/nandita005/decodex_final/blob/master/src/config.py"
click node_risk "https://github.com/nandita005/decodex_final/blob/master/src/risk_scoring/client_risk.py"
click node_classifier "https://github.com/nandita005/decodex_final/blob/master/src/classification/mechanism_classifier.py"
click node_evidence "https://github.com/nandita005/decodex_final/blob/master/src/explainability/evidence_generator.py"
click node_integrate "https://github.com/nandita005/decodex_final/blob/master/src/s3/integrate_evidence.py"
click node_composite "https://github.com/nandita005/decodex_final/blob/master/src/s3/compute_composite_risk.py"
click node_escalation "https://github.com/nandita005/decodex_final/blob/master/src/s3/select_escalation.py"
click node_explanations "https://github.com/nandita005/decodex_final/blob/master/src/s3/generate_explanations.py"
click node_outputs "https://github.com/nandita005/decodex_final/tree/master/outputs"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_loader,node_trade_clean,node_orders_clean,node_batch toneBlue
class node_trade_graph,node_cycle,node_centrality,node_network_extensions,node_micro,node_raw,node_outputs toneAmber
class node_fusion,node_config,node_risk,node_classifier toneMint
class node_evidence,node_integrate,node_composite,node_escalation,node_explanations toneRose
class node_analyst toneIndigo
class node_runner toneTeal
```

## Components

### Data preparation

- [Load and clean](https://github.com/nandita005/decodex_final/blob/master/src/preprocessing/load_data.py): `src/preprocessing/load_data.py`
- [Trade cleaning](https://github.com/nandita005/decodex_final/blob/master/src/preprocessing/clean_trades.py): `src/preprocessing/clean_trades.py`
- [Order cleaning](https://github.com/nandita005/decodex_final/blob/master/src/preprocessing/clean_orders.py): `src/preprocessing/clean_orders.py`
- [Scrip/date batches](https://github.com/nandita005/decodex_final/blob/master/src/preprocessing/build_scrip_date_batches.py): `src/preprocessing/build_scrip_date_batches.py`

### Detection signals

- [Trade graph](https://github.com/nandita005/decodex_final/blob/master/src/network_analysis/build_trade_graph.py): `src/network_analysis/build_trade_graph.py`
- [Cycle detection](https://github.com/nandita005/decodex_final/blob/master/src/network_analysis/cycle_detection.py): `src/network_analysis/cycle_detection.py`
- [Centrality and reciprocity](https://github.com/nandita005/decodex_final/blob/master/src/network_analysis/centrality_metrics.py): `src/network_analysis/centrality_metrics.py`
- [Network analysis](https://github.com/nandita005/decodex_final/tree/master/src/network_analysis): `src/network_analysis`
- [Microstructure features](https://github.com/nandita005/decodex_final/blob/master/src/microstructure/price_volume.py): `src/microstructure/price_volume.py`

### Risk assessment

- [Signal fusion](https://github.com/nandita005/decodex_final/blob/master/src/reasoning_engine/signal_fusion.py): `src/reasoning_engine/signal_fusion.py`
- [Risk configuration](https://github.com/nandita005/decodex_final/blob/master/src/config.py): `src/config.py`
- [Client and member risk](https://github.com/nandita005/decodex_final/blob/master/src/risk_scoring/client_risk.py): `src/risk_scoring/client_risk.py`
- [Mechanism labels](https://github.com/nandita005/decodex_final/blob/master/src/classification/mechanism_classifier.py): `src/classification/mechanism_classifier.py`

### Evidence and review

- [Evidence records](https://github.com/nandita005/decodex_final/blob/master/src/explainability/evidence_generator.py): `src/explainability/evidence_generator.py`
- [Evidence integration](https://github.com/nandita005/decodex_final/blob/master/src/s3/integrate_evidence.py): `src/s3/integrate_evidence.py`
- [Composite scoring](https://github.com/nandita005/decodex_final/blob/master/src/s3/compute_composite_risk.py): `src/s3/compute_composite_risk.py`
- [Escalation selection](https://github.com/nandita005/decodex_final/blob/master/src/s3/select_escalation.py): `src/s3/select_escalation.py`
- [Review explanations](https://github.com/nandita005/decodex_final/blob/master/src/s3/generate_explanations.py): `src/s3/generate_explanations.py`

### Other components

- **Analyst**
- [Raw market files](https://github.com/nandita005/decodex_final/tree/master/data/raw): `data/raw`
- [Pipeline runner](https://github.com/nandita005/decodex_final/blob/master/run_pipeline.py): `run_pipeline.py`
- [Reports and graphs](https://github.com/nandita005/decodex_final/tree/master/outputs): `outputs`

## Connections

- Raw market files → Load and clean: loads files (evidence: `README.md`)
- Pipeline runner → Load and clean: loads data (evidence: `run_pipeline.py`)
- Load and clean → Trade cleaning: cleans trades (evidence: `src/preprocessing/load_data.py`)
- Load and clean → Order cleaning: cleans orders (evidence: `src/preprocessing/load_data.py`)
- Pipeline runner → Scrip/date batches: builds batches (evidence: `run_pipeline.py`)
- Pipeline runner → Trade graph: builds graph (evidence: `run_pipeline.py`)
- Trade graph → Trade cleaning: normalizes IDs (evidence: `src/network_analysis/build_trade_graph.py`)
- Pipeline runner → Cycle detection: detects cycles (evidence: `run_pipeline.py`)
- Pipeline runner → Centrality and reciprocity: computes metrics (evidence: `run_pipeline.py`)
- Pipeline runner → Microstructure features: computes features (evidence: `run_pipeline.py`)
- Pipeline runner → Signal fusion: fuses signals (evidence: `run_pipeline.py`)
- Signal fusion → Centrality and reciprocity: uses metrics (evidence: `README.md`)
- Signal fusion → Microstructure features: uses features (evidence: `README.md`)
- Signal fusion → Cycle detection: uses cycle nodes (evidence: `README.md`)
- Pipeline runner → Client and member risk: scores clients (evidence: `run_pipeline.py`)
- Client and member risk → Signal fusion: reads components (evidence: `src/risk_scoring/client_risk.py`)
- Client and member risk → Risk configuration: validates weights (evidence: `src/risk_scoring/client_risk.py`)
- Pipeline runner → Mechanism labels: classifies clients (evidence: `run_pipeline.py`)
- Pipeline runner → Evidence records: generates evidence (evidence: `run_pipeline.py`)
- Evidence records → Signal fusion: uses signal names (evidence: `src/explainability/evidence_generator.py`)
- Pipeline runner → Evidence integration: runs stage three (evidence: `run_pipeline.py`)
- Pipeline runner → Composite scoring: computes final risk (evidence: `run_pipeline.py`)
- Pipeline runner → Escalation selection: selects escalation (evidence: `run_pipeline.py`)
- Pipeline runner → Review explanations: writes explanations (evidence: `run_pipeline.py`)
- Evidence integration → Reports and graphs: writes combined records (evidence: `src/s3/integrate_evidence.py`)
- Composite scoring → Reports and graphs: writes scores (evidence: `src/s3/compute_composite_risk.py`)
- Review explanations → Reports and graphs: writes report (evidence: `src/s3/generate_explanations.py`)
- Pipeline runner → Reports and graphs: writes artifacts (evidence: `README.md`)
- Trade graph → Reports and graphs: renders graphs (evidence: `README.md`)
- Network analysis → Analyst: supports review
- Reports and graphs → Analyst: provides review artifacts (evidence: `README.md`)
