# deeepikashri/credit-risk-ml architecture

> Architecture diagram of the GitHub repository deeepikashri/credit-risk-ml, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-07). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/deeepikashri/credit-risk-ml
- Repository: https://github.com/deeepikashri/credit-risk-ml

## Overview

This repository is a command-line credit-default classifier: training generates and splits data, builds and evaluates a scikit-learn pipeline, optionally tunes it, reports metrics and feature importance, and saves the fitted pipeline. Prediction loads that artifact and returns a default probability and risk label for supplied customer features. The README documents the overall workflow and outputs; source excerpts show the training and prediction calls. The tree does not include the documented src/credit_risk layout, so paths below use the actual root-level files.

Main entry points: [`train.py`](https://github.com/deeepikashri/credit-risk-ml/blob/main/train.py) (Training orchestration), [`predict.py`](https://github.com/deeepikashri/credit-risk-ml/blob/main/predict.py) (Prediction interface), [`pipeline.py`](https://github.com/deeepikashri/credit-risk-ml/blob/main/pipeline.py) (Preprocessing and forest), [`config.py`](https://github.com/deeepikashri/credit-risk-ml/blob/main/config.py) (Feature configuration)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_data["Data and configuration"]
  node_config["Feature configuration<br/>[config.py]"]
  node_synthetic_data["Synthetic data<br/>[data.py]"]
end

subgraph group_model["Model pipeline"]
  node_preprocessing["Preprocessing and forest<br/>[pipeline.py]"]
  node_search["Hyperparameter search<br/>[pipeline.py]"]
end

subgraph group_training["Training and analysis"]
  node_training_cli["Training orchestration<br/>[train.py]"]
  node_cross_validation["Cross-validation<br/>[train.py]"]
  node_evaluation["Evaluation metrics<br/>[evaluate.py]"]
  node_explainability["Feature importance<br/>[explain.py]"]
  node_reports["Metrics and plots<br/>[metrics.json]"]
end

subgraph group_inference["Prediction"]
  node_model_artifact["Serialized pipeline"]
  node_prediction_cli["Prediction interface<br/>[predict.py]"]
  node_risk_scoring["Risk scoring<br/>[predict.py]"]
end

node_operator(("CLI user"))
node_customer_data["Customer features"]

node_operator -->|"runs training"| node_training_cli
node_training_cli -->|"generates and splits"| node_synthetic_data
node_synthetic_data -->|"uses feature definitions"| node_config
node_training_cli -->|"builds and fits"| node_preprocessing
node_preprocessing -->|"uses feature lists"| node_config
node_training_cli -->|"scores folds"| node_cross_validation
node_training_cli -->|"optionally tunes"| node_search
node_search -->|"searches pipeline parameters"| node_preprocessing
node_training_cli -->|"computes metrics"| node_evaluation
node_training_cli -->|"computes importance"| node_explainability
node_training_cli -->|"serializes fitted model"| node_model_artifact
node_training_cli -->|"writes metrics and plots"| node_reports
node_operator -->|"runs prediction"| node_prediction_cli
node_customer_data -->|"provides features"| node_prediction_cli
node_prediction_cli -->|"loads model"| node_model_artifact
node_prediction_cli -->|"scores customers"| node_risk_scoring
node_risk_scoring -->|"uses features and threshold"| node_config

click node_config "https://github.com/deeepikashri/credit-risk-ml/blob/main/config.py"
click node_synthetic_data "https://github.com/deeepikashri/credit-risk-ml/blob/main/data.py"
click node_preprocessing "https://github.com/deeepikashri/credit-risk-ml/blob/main/pipeline.py"
click node_search "https://github.com/deeepikashri/credit-risk-ml/blob/main/pipeline.py"
click node_training_cli "https://github.com/deeepikashri/credit-risk-ml/blob/main/train.py"
click node_cross_validation "https://github.com/deeepikashri/credit-risk-ml/blob/main/train.py"
click node_evaluation "https://github.com/deeepikashri/credit-risk-ml/blob/main/evaluate.py"
click node_explainability "https://github.com/deeepikashri/credit-risk-ml/blob/main/explain.py"
click node_reports "https://github.com/deeepikashri/credit-risk-ml/blob/main/metrics.json"
click node_prediction_cli "https://github.com/deeepikashri/credit-risk-ml/blob/main/predict.py"
click node_risk_scoring "https://github.com/deeepikashri/credit-risk-ml/blob/main/predict.py"

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_config,node_synthetic_data,node_operator toneBlue
class node_preprocessing,node_search toneAmber
class node_training_cli,node_cross_validation,node_evaluation,node_explainability,node_reports toneMint
class node_model_artifact,node_prediction_cli,node_risk_scoring toneRose
class node_customer_data toneIndigo
```

## Components

### Data and configuration

- [Feature configuration](https://github.com/deeepikashri/credit-risk-ml/blob/main/config.py): `config.py`
- [Synthetic data](https://github.com/deeepikashri/credit-risk-ml/blob/main/data.py): `data.py`

### Model pipeline

- [Preprocessing and forest](https://github.com/deeepikashri/credit-risk-ml/blob/main/pipeline.py): `pipeline.py`
- [Hyperparameter search](https://github.com/deeepikashri/credit-risk-ml/blob/main/pipeline.py): `pipeline.py`

### Training and analysis

- [Training orchestration](https://github.com/deeepikashri/credit-risk-ml/blob/main/train.py): `train.py`
- [Cross-validation](https://github.com/deeepikashri/credit-risk-ml/blob/main/train.py): `train.py`
- [Evaluation metrics](https://github.com/deeepikashri/credit-risk-ml/blob/main/evaluate.py): `evaluate.py`
- [Feature importance](https://github.com/deeepikashri/credit-risk-ml/blob/main/explain.py): `explain.py`
- [Metrics and plots](https://github.com/deeepikashri/credit-risk-ml/blob/main/metrics.json): `metrics.json`

### Prediction

- **Serialized pipeline**
- [Prediction interface](https://github.com/deeepikashri/credit-risk-ml/blob/main/predict.py): `predict.py`
- [Risk scoring](https://github.com/deeepikashri/credit-risk-ml/blob/main/predict.py): `predict.py`

### Other components

- **CLI user**
- **Customer features**

## Connections

- CLI user → Training orchestration: runs training (evidence: `README.md`)
- Training orchestration → Synthetic data: generates and splits (evidence: `train.py`)
- Synthetic data → Feature configuration: uses feature definitions (evidence: `data.py`)
- Training orchestration → Preprocessing and forest: builds and fits (evidence: `train.py`)
- Preprocessing and forest → Feature configuration: uses feature lists (evidence: `pipeline.py`)
- Training orchestration → Cross-validation: scores folds (evidence: `train.py`)
- Training orchestration → Hyperparameter search: optionally tunes (evidence: `train.py`)
- Hyperparameter search → Preprocessing and forest: searches pipeline parameters (evidence: `pipeline.py`)
- Training orchestration → Evaluation metrics: computes metrics (evidence: `train.py`)
- Training orchestration → Feature importance: computes importance (evidence: `train.py`)
- Training orchestration → Serialized pipeline: serializes fitted model (evidence: `train.py`)
- Training orchestration → Metrics and plots: writes metrics and plots (evidence: `train.py`)
- CLI user → Prediction interface: runs prediction (evidence: `README.md`)
- Customer features → Prediction interface: provides features (evidence: `README.md`)
- Prediction interface → Serialized pipeline: loads model (evidence: `predict.py`)
- Prediction interface → Risk scoring: scores customers (evidence: `predict.py`)
- Risk scoring → Feature configuration: uses features and threshold (evidence: `predict.py`)
