# jay-feng2008/phonocardiogram-classification architecture

> Architecture diagram of the GitHub repository jay-feng2008/phonocardiogram-classification, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-03). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/jay-feng2008/phonocardiogram-classification
- Repository: https://github.com/jay-feng2008/phonocardiogram-classification

## Overview

This repository is a research implementation and experiment archive for phonocardiogram-based valvular heart disease classification, rather than a packaged application. The documented workflow is to train a ProbSparse self-attention classifier with virtual adversarial training, tune it with Bayesian optimization, and evaluate classification performance. The tree also preserves cross-validation, baseline/model comparisons, and ablation results. Source excerpts confirm the TensorFlow training and attention-layer components; several data-preparation and evaluation connections are documented only broadly, so those links are marked unverified or omitted.

Main entry points: [`metrics`](https://github.com/jay-feng2008/phonocardiogram-classification/tree/main/metrics) (Model Metrics), [`layers.py`](https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/layers.py) (ProbSparse Layers), [`Config-Oct-9/main.py`](https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/Config-Oct-9/main.py) (VAT Training), [`Config-Oct-9/10-fold-crossvalidation/main.py`](https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/Config-Oct-9/10-fold-crossvalidation/main.py) (10-Fold Evaluation)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_data["PCG Data"]
  node_raw[("Raw PCG Data")]
  node_features[("MFCC Features<br/>[mfcc.npz]")]
end

subgraph group_model["Classification Model"]
  node_attention["ProbSparse Layers<br/>[layers.py]"]
  node_encoder["Transformer Encoder<br/>[encoder.py]"]
end

subgraph group_experiments["Training Experiments"]
  node_main_training["VAT Training<br/>[main.py]"]
  node_vat_variant["VAT Baseline"]
  node_crossval["10-Fold Evaluation<br/>[main.py]"]
  node_nov_crossval["Updated Cross-Validation"]
  node_bayes["Bayesian Tuning<br/>[main.py]"]
  node_gridsearch["Hyperparameter Search"]
  node_baseline["Baseline Training<br/>[model_baseline.py]"]
  node_inception["InceptionTime Comparison"]
  node_mlstm["MLSTM-FCN Comparison"]
  node_resnet["ResNet Comparison"]
end

subgraph group_evaluation["Evaluation Results"]
  node_ablation["Ablation Study"]
  node_metrics["Model Metrics"]
  node_figures["Result Figures"]
end

node_researcher(("Researcher"))

node_researcher -.->|"runs"| node_main_training
node_raw -.->|"feature extraction"| node_features
node_features -.->|"trains on"| node_main_training
node_main_training -->|"uses layers"| node_attention
node_vat_variant -->|"uses layers"| node_attention
node_crossval -->|"uses layers"| node_attention
node_nov_crossval -->|"uses layers"| node_attention
node_bayes -->|"uses layers"| node_attention
node_baseline -->|"uses layers"| node_attention
node_bayes -.->|"tunes training"| node_main_training
node_main_training -.->|"evaluates"| node_metrics
node_crossval -.->|"reports results"| node_metrics
node_ablation -.->|"compares components"| node_metrics
node_inception -.->|"model comparison"| node_metrics
node_mlstm -.->|"model comparison"| node_metrics
node_resnet -.->|"model comparison"| node_metrics
node_metrics -.->|"summarized in"| node_figures

click node_raw "https://github.com/jay-feng2008/phonocardiogram-classification/tree/main/raw_data"
click node_features "https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/Config-Oct-9/model_baseline/mfcc.npz"
click node_main_training "https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/Config-Oct-9/main.py"
click node_vat_variant "https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/Config-Oct-9/VAT/model_baseline%2Bvat.py"
click node_crossval "https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/Config-Oct-9/10-fold-crossvalidation/main.py"
click node_nov_crossval "https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/Config-Nov-17-2023/10-fold-cross-validation.py"
click node_bayes "https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/Config-Oct-9/Bayesian_optimization/main.py"
click node_gridsearch "https://github.com/jay-feng2008/phonocardiogram-classification/tree/main/Config-Oct-9/grid_search"
click node_baseline "https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/Config-Oct-9/model_baseline/model_baseline.py"
click node_attention "https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/layers.py"
click node_encoder "https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/Config-Oct-9/encoder.py"
click node_inception "https://github.com/jay-feng2008/phonocardiogram-classification/tree/main/InceptionTime"
click node_mlstm "https://github.com/jay-feng2008/phonocardiogram-classification/tree/main/MLSTM-FCN"
click node_resnet "https://github.com/jay-feng2008/phonocardiogram-classification/tree/main/Resnet"
click node_ablation "https://github.com/jay-feng2008/phonocardiogram-classification/tree/main/Ablation%20Study"
click node_metrics "https://github.com/jay-feng2008/phonocardiogram-classification/tree/main/metrics"
click node_figures "https://github.com/jay-feng2008/phonocardiogram-classification/tree/main/Figures"

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_raw,node_features toneBlue
class node_attention,node_encoder toneAmber
class node_main_training,node_vat_variant,node_crossval,node_nov_crossval,node_bayes,node_gridsearch,node_baseline,node_inception,node_mlstm,node_resnet toneMint
class node_ablation,node_metrics,node_figures toneRose
class node_researcher toneIndigo
```

## Components

### PCG Data

- [Raw PCG Data](https://github.com/jay-feng2008/phonocardiogram-classification/tree/main/raw_data): `raw_data`
- [MFCC Features](https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/Config-Oct-9/model_baseline/mfcc.npz): `Config-Oct-9/model_baseline/mfcc.npz`

### Classification Model

- [ProbSparse Layers](https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/layers.py): `layers.py`
- [Transformer Encoder](https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/Config-Oct-9/encoder.py): `Config-Oct-9/encoder.py`

### Training Experiments

- [VAT Training](https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/Config-Oct-9/main.py): `Config-Oct-9/main.py`
- [VAT Baseline](https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/Config-Oct-9/VAT/model_baseline%2Bvat.py): `Config-Oct-9/VAT/model_baseline+vat.py`
- [10-Fold Evaluation](https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/Config-Oct-9/10-fold-crossvalidation/main.py): `Config-Oct-9/10-fold-crossvalidation/main.py`
- [Updated Cross-Validation](https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/Config-Nov-17-2023/10-fold-cross-validation.py): `Config-Nov-17-2023/10-fold-cross-validation.py`
- [Bayesian Tuning](https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/Config-Oct-9/Bayesian_optimization/main.py): `Config-Oct-9/Bayesian_optimization/main.py`
- [Hyperparameter Search](https://github.com/jay-feng2008/phonocardiogram-classification/tree/main/Config-Oct-9/grid_search): `Config-Oct-9/grid_search`
- [Baseline Training](https://github.com/jay-feng2008/phonocardiogram-classification/blob/main/Config-Oct-9/model_baseline/model_baseline.py): `Config-Oct-9/model_baseline/model_baseline.py`
- [InceptionTime Comparison](https://github.com/jay-feng2008/phonocardiogram-classification/tree/main/InceptionTime): `InceptionTime`
- [MLSTM-FCN Comparison](https://github.com/jay-feng2008/phonocardiogram-classification/tree/main/MLSTM-FCN): `MLSTM-FCN`
- [ResNet Comparison](https://github.com/jay-feng2008/phonocardiogram-classification/tree/main/Resnet): `Resnet`

### Evaluation Results

- [Ablation Study](https://github.com/jay-feng2008/phonocardiogram-classification/tree/main/Ablation%20Study): `Ablation Study`
- [Model Metrics](https://github.com/jay-feng2008/phonocardiogram-classification/tree/main/metrics): `metrics`
- [Result Figures](https://github.com/jay-feng2008/phonocardiogram-classification/tree/main/Figures): `Figures`

### Other components

- **Researcher**

## Connections

- Researcher → VAT Training: runs
- Raw PCG Data → MFCC Features: feature extraction
- MFCC Features → VAT Training: trains on
- VAT Training → ProbSparse Layers: uses layers (evidence: `Config-Oct-9/main.py`)
- VAT Baseline → ProbSparse Layers: uses layers (evidence: `Config-Oct-9/VAT/model_baseline+vat.py`)
- 10-Fold Evaluation → ProbSparse Layers: uses layers (evidence: `Config-Oct-9/10-fold-crossvalidation/main.py`)
- Updated Cross-Validation → ProbSparse Layers: uses layers (evidence: `Config-Nov-17-2023/10-fold-cross-validation.py`)
- Bayesian Tuning → ProbSparse Layers: uses layers (evidence: `Config-Oct-9/Bayesian_optimization/main.py`)
- Baseline Training → ProbSparse Layers: uses layers (evidence: `Config-Oct-9/model_baseline/model_baseline.py`)
- Bayesian Tuning → VAT Training: tunes training (evidence: `README.md`)
- VAT Training → Model Metrics: evaluates (evidence: `README.md`)
- 10-Fold Evaluation → Model Metrics: reports results
- Ablation Study → Model Metrics: compares components (evidence: `README.md`)
- InceptionTime Comparison → Model Metrics: model comparison
- MLSTM-FCN Comparison → Model Metrics: model comparison
- ResNet Comparison → Model Metrics: model comparison
- Model Metrics → Result Figures: summarized in
