# kryp71cgh057/auto-encoder-model architecture

> Architecture diagram of the GitHub repository kryp71cgh057/auto-encoder-model, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-11-11). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/kryp71cgh057/auto-encoder-model
- Repository: https://github.com/kryp71cgh057/auto-encoder-model

## Overview

This repository is a research notebook for unsupervised financial-fraud detection, not a reusable multi-service application. Its documented workflow uses transaction data to train an auto-encoder on legitimate transactions, flags anomalies using reconstruction error and a percentile threshold, and evaluates results against fraud labels. The README also describes baseline comparison and generated reports, plots, and model artifacts. The notebook is present in the tree but its source was not excerpted, so component-level implementation details and relationships rely on README documentation.

Main entry points: [`auto_encoder_fraud_detection.ipynb`](https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb) (Metric evaluation), [`comprehensive_report_20260320_165611.txt`](https://github.com/kryp71cgh057/auto-encoder-model/blob/main/comprehensive_report_20260320_165611.txt) (Comprehensive report)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_data["Data preparation"]
  node_prepare["Feature preparation"]
  node_partition["Data partitioning"]
end

subgraph group_detection["Anomaly detection"]
  node_train["Legitimate-only training"]
  node_autoencoder["Auto-encoder"]
  node_reconstruction["Reconstruction error"]
  node_threshold["Anomaly threshold"]
  node_risk["Risk tiering"]
end

subgraph group_evaluation["Evaluation and outputs"]
  node_evaluation["Metric evaluation"]
  node_baseline["Baseline comparison"]
  node_visuals["Diagnostic visualizations"]
  node_riskreport["Fraud risk report"]
  node_modeloutput["Saved model and config"]
  node_report["Comprehensive report"]
end

node_analyst(("Researcher"))
node_dataset["PaySim dataset"]

node_analyst -->|"supplies data"| node_dataset
node_dataset -->|"provides transactions"| node_prepare
node_prepare -->|"prepares features"| node_partition
node_partition -->|"selects legitimate rows"| node_train
node_train -->|"trains model"| node_autoencoder
node_partition -->|"provides evaluation data"| node_autoencoder
node_autoencoder -->|"reconstructs inputs"| node_reconstruction
node_reconstruction -->|"compares errors"| node_threshold
node_threshold -->|"assigns risk tier"| node_risk
node_risk -->|"produces report"| node_riskreport
node_risk -->|"evaluates predictions"| node_evaluation
node_evaluation -->|"compares baseline"| node_baseline
node_evaluation -->|"generates plots"| node_visuals
node_evaluation -->|"generates report"| node_report
node_autoencoder -->|"saves model"| node_modeloutput

click node_prepare "https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb"
click node_partition "https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb"
click node_train "https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb"
click node_autoencoder "https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb"
click node_reconstruction "https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb"
click node_threshold "https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb"
click node_risk "https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb"
click node_evaluation "https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb"
click node_baseline "https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb"
click node_visuals "https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb"
click node_riskreport "https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb"
click node_modeloutput "https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb"
click node_report "https://github.com/kryp71cgh057/auto-encoder-model/blob/main/comprehensive_report_20260320_165611.txt"

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_prepare,node_partition toneBlue
class node_train,node_autoencoder,node_reconstruction,node_threshold,node_risk toneAmber
class node_evaluation,node_baseline,node_visuals,node_riskreport,node_modeloutput,node_report toneMint
class node_analyst,node_dataset toneIndigo
```

## Components

### Data preparation

- [Feature preparation](https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb): `auto_encoder_fraud_detection.ipynb`
- [Data partitioning](https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb): `auto_encoder_fraud_detection.ipynb`

### Anomaly detection

- [Legitimate-only training](https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb): `auto_encoder_fraud_detection.ipynb`
- [Auto-encoder](https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb): `auto_encoder_fraud_detection.ipynb`
- [Reconstruction error](https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb): `auto_encoder_fraud_detection.ipynb`
- [Anomaly threshold](https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb): `auto_encoder_fraud_detection.ipynb`
- [Risk tiering](https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb): `auto_encoder_fraud_detection.ipynb`

### Evaluation and outputs

- [Metric evaluation](https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb): `auto_encoder_fraud_detection.ipynb`
- [Baseline comparison](https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb): `auto_encoder_fraud_detection.ipynb`
- [Diagnostic visualizations](https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb): `auto_encoder_fraud_detection.ipynb`
- [Fraud risk report](https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb): `auto_encoder_fraud_detection.ipynb`
- [Saved model and config](https://github.com/kryp71cgh057/auto-encoder-model/blob/main/auto_encoder_fraud_detection.ipynb): `auto_encoder_fraud_detection.ipynb`
- [Comprehensive report](https://github.com/kryp71cgh057/auto-encoder-model/blob/main/comprehensive_report_20260320_165611.txt): `comprehensive_report_20260320_165611.txt`

### Other components

- **Researcher**
- **PaySim dataset**

## Connections

- Researcher → PaySim dataset: supplies data (evidence: `README.md`)
- PaySim dataset → Feature preparation: provides transactions (evidence: `README.md`)
- Feature preparation → Data partitioning: prepares features (evidence: `README.md`)
- Data partitioning → Legitimate-only training: selects legitimate rows (evidence: `README.md`)
- Legitimate-only training → Auto-encoder: trains model (evidence: `README.md`)
- Data partitioning → Auto-encoder: provides evaluation data (evidence: `README.md`)
- Auto-encoder → Reconstruction error: reconstructs inputs (evidence: `README.md`)
- Reconstruction error → Anomaly threshold: compares errors (evidence: `README.md`)
- Anomaly threshold → Risk tiering: assigns risk tier (evidence: `README.md`)
- Risk tiering → Fraud risk report: produces report (evidence: `README.md`)
- Risk tiering → Metric evaluation: evaluates predictions (evidence: `README.md`)
- Metric evaluation → Baseline comparison: compares baseline (evidence: `README.md`)
- Metric evaluation → Diagnostic visualizations: generates plots (evidence: `README.md`)
- Metric evaluation → Comprehensive report: generates report (evidence: `README.md`)
- Auto-encoder → Saved model and config: saves model (evidence: `README.md`)