# peerahammad/pard-in-network-traffic-using-switching-state-space-models architecture

> Architecture diagram of the GitHub repository peerahammad/pard-in-network-traffic-using-switching-state-space-models, 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/peerahammad/pard-in-network-traffic-using-switching-state-space-models
- Repository: https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models

## Overview

PARD-SSM is a research system for inferring changing network-attack regimes and hidden states from traffic sequences. The documented workflow loads network data, engineers normalized and windowed observations, applies state-space models and inference methods, then evaluates and visualizes regime estimates. The graph preserves the documented explainability and experiment capabilities. Several links reflect README-level architecture because their implementation details were not sampled; those are cited to the README rather than inferred from module co-location.

Main entry points: [`experiments/run_baseline.py`](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/experiments/run_baseline.py) (Baseline experiments), [`src/data_processing/feature_engineering.py`](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/data_processing/feature_engineering.py) (Feature engineering), [`experiments/run_switching.py`](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/experiments/run_switching.py) (Switching experiments), [`src/models/switching_ssm.py`](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/models/switching_ssm.py) (Switching SSM)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_data["Traffic preparation"]
  node_loader["Dataset loader<br/>[dataset_loader.py]"]
  node_features["Feature engineering"]
end

subgraph group_models["State-space models"]
  node_linear["Linear SSM<br/>[linear_ssm.py]"]
  node_nonlinear["Nonlinear SSM<br/>[nonlinear_ssm.py]"]
  node_switching["Switching SSM<br/>[switching_ssm.py]"]
end

subgraph group_inference["Inference methods"]
  node_kf["Kalman filter<br/>[kalman_filter.py]"]
  node_ekf["Extended Kalman filter<br/>[ekf.py]"]
  node_ukf["Unscented Kalman filter<br/>[ukf.py]"]
  node_variational["Variational switching"]
end

subgraph group_analysis["Evaluation and analysis"]
  node_metrics["Regime metrics<br/>[metrics.py]"]
  node_visualization["Result visualization<br/>[visualization.py]"]
  node_shap["SHAP explainability<br/>[shap_explainer.py]"]
end

subgraph group_experiments["Research experiments"]
  node_baseline_runner["Baseline experiments<br/>[run_baseline.py]"]
  node_switch_runner["Switching experiments<br/>[run_switching.py]"]
  node_baselines["Baseline comparisons<br/>[baselines.py]"]
  node_ablation["Ablation studies<br/>[ablation_study.py]"]
  node_evaluation["Evaluation metrics"]
  node_significance["Significance analysis"]
end

node_traffic(("Network traffic"))

node_traffic -->|"loads data"| node_loader
node_loader -->|"provides datasets"| node_features
node_features -->|"builds observations"| node_linear
node_features -->|"builds observations"| node_switching
node_linear -->|"builds filter"| node_kf
node_nonlinear -->|"builds filter"| node_ekf
node_nonlinear -->|"builds filter"| node_ukf
node_switching -->|"infers regimes"| node_variational
node_baseline_runner -->|"preprocesses data"| node_features
node_baseline_runner -->|"runs baselines"| node_linear
node_baseline_runner -->|"runs baselines"| node_nonlinear
node_baseline_runner -->|"computes metrics"| node_metrics
node_baseline_runner -->|"plots results"| node_visualization
node_switch_runner -->|"preprocesses data"| node_features
node_switch_runner -->|"runs inference"| node_variational
node_switch_runner -->|"computes metrics"| node_metrics
node_switch_runner -->|"plots results"| node_visualization
node_baselines -.->|"compares methods"| node_baseline_runner
node_ablation -.->|"tests variants"| node_switching
node_evaluation -->|"uses metrics"| node_metrics
node_evaluation -->|"plots evaluations"| node_visualization
node_significance -.->|"tests results"| node_evaluation
node_switching -.->|"supports explanations"| node_shap

click node_loader "https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/data_processing/dataset_loader.py"
click node_features "https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/data_processing/feature_engineering.py"
click node_linear "https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/models/linear_ssm.py"
click node_nonlinear "https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/models/nonlinear_ssm.py"
click node_switching "https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/models/switching_ssm.py"
click node_kf "https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/inference/kalman_filter.py"
click node_ekf "https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/inference/ekf.py"
click node_ukf "https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/inference/ukf.py"
click node_variational "https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/inference/variational_switching.py"
click node_metrics "https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/utils/metrics.py"
click node_visualization "https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/utils/visualization.py"
click node_shap "https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/explainablity/shap_explainer.py"
click node_baseline_runner "https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/experiments/run_baseline.py"
click node_switch_runner "https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/experiments/run_switching.py"
click node_baselines "https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/experiments/baselines.py"
click node_ablation "https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/experiments/ablation_study.py"
click node_evaluation "https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/experiments/evaluation_metrics.py"
click node_significance "https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/experiments/statistical_significance.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_loader,node_features toneBlue
class node_linear,node_nonlinear,node_switching toneAmber
class node_kf,node_ekf,node_ukf,node_variational toneMint
class node_metrics,node_visualization,node_shap toneRose
class node_baseline_runner,node_switch_runner,node_baselines,node_ablation,node_evaluation,node_significance,node_traffic toneIndigo
```

## Components

### Traffic preparation

- [Dataset loader](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/data_processing/dataset_loader.py): `src/data_processing/dataset_loader.py`
- [Feature engineering](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/data_processing/feature_engineering.py): `src/data_processing/feature_engineering.py`

### State-space models

- [Linear SSM](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/models/linear_ssm.py): `src/models/linear_ssm.py`
- [Nonlinear SSM](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/models/nonlinear_ssm.py): `src/models/nonlinear_ssm.py`
- [Switching SSM](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/models/switching_ssm.py): `src/models/switching_ssm.py`

### Inference methods

- [Kalman filter](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/inference/kalman_filter.py): `src/inference/kalman_filter.py`
- [Extended Kalman filter](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/inference/ekf.py): `src/inference/ekf.py`
- [Unscented Kalman filter](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/inference/ukf.py): `src/inference/ukf.py`
- [Variational switching](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/inference/variational_switching.py): `src/inference/variational_switching.py`

### Evaluation and analysis

- [Regime metrics](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/utils/metrics.py): `src/utils/metrics.py`
- [Result visualization](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/utils/visualization.py): `src/utils/visualization.py`
- [SHAP explainability](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/src/explainablity/shap_explainer.py): `src/explainablity/shap_explainer.py`

### Research experiments

- [Baseline experiments](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/experiments/run_baseline.py): `experiments/run_baseline.py`
- [Switching experiments](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/experiments/run_switching.py): `experiments/run_switching.py`
- [Baseline comparisons](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/experiments/baselines.py): `experiments/baselines.py`
- [Ablation studies](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/experiments/ablation_study.py): `experiments/ablation_study.py`
- [Evaluation metrics](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/experiments/evaluation_metrics.py): `experiments/evaluation_metrics.py`
- [Significance analysis](https://github.com/peerahammad/pard-in-network-traffic-using-switching-state-space-models/blob/main/experiments/statistical_significance.py): `experiments/statistical_significance.py`

### Other components

- **Network traffic**

## Connections

- Network traffic → Dataset loader: loads data (evidence: `README.md`)
- Dataset loader → Feature engineering: provides datasets (evidence: `README.md`)
- Feature engineering → Linear SSM: builds observations (evidence: `README.md`)
- Feature engineering → Switching SSM: builds observations (evidence: `README.md`)
- Linear SSM → Kalman filter: builds filter (evidence: `src/models/linear_ssm.py`)
- Nonlinear SSM → Extended Kalman filter: builds filter (evidence: `src/models/nonlinear_ssm.py`)
- Nonlinear SSM → Unscented Kalman filter: builds filter (evidence: `src/models/nonlinear_ssm.py`)
- Switching SSM → Variational switching: infers regimes (evidence: `README.md`)
- Baseline experiments → Feature engineering: preprocesses data (evidence: `experiments/run_baseline.py`)
- Baseline experiments → Linear SSM: runs baselines (evidence: `experiments/run_baseline.py`)
- Baseline experiments → Nonlinear SSM: runs baselines (evidence: `experiments/run_baseline.py`)
- Baseline experiments → Regime metrics: computes metrics (evidence: `experiments/run_baseline.py`)
- Baseline experiments → Result visualization: plots results (evidence: `experiments/run_baseline.py`)
- Switching experiments → Feature engineering: preprocesses data (evidence: `experiments/run_switching.py`)
- Switching experiments → Variational switching: runs inference (evidence: `experiments/run_switching.py`)
- Switching experiments → Regime metrics: computes metrics (evidence: `experiments/run_switching.py`)
- Switching experiments → Result visualization: plots results (evidence: `experiments/run_switching.py`)
- Baseline comparisons → Baseline experiments: compares methods
- Ablation studies → Switching SSM: tests variants
- Evaluation metrics → Regime metrics: uses metrics (evidence: `experiments/evaluation_metrics.py`)
- Evaluation metrics → Result visualization: plots evaluations (evidence: `experiments/evaluation_metrics.py`)
- Significance analysis → Evaluation metrics: tests results
- Switching SSM → SHAP explainability: supports explanations (evidence: `README.md`)
