# ashwinrj/federated-learning-pytorch architecture

> Architecture diagram of the GitHub repository ashwinrj/federated-learning-pytorch, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-04). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/ashwinrj/federated-learning-pytorch
- Repository: https://github.com/ashwinrj/federated-learning-pytorch

## Overview

This PyTorch research implementation runs conventional baseline training or federated experiments on MNIST, Fashion-MNIST, and CIFAR-10. The federated workflow loads and partitions training data, selects clients each round, trains local models, averages their weights into a global model, and evaluates it; baseline training follows a separate single-model path. Both paths share model definitions and evaluation support. Dataset acquisition and experiment outputs are shown where source or README evidence supports them; source excerpts do not establish every runtime detail.

Main entry points: [`src/federated_main.py`](https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/federated_main.py) (Federated coordinator), [`src/baseline_main.py`](https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/baseline_main.py) (Baseline trainer), [`src/utils.py`](https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/utils.py) (Dataset loader), [`src/update.py`](https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/update.py) (Local training)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_experiments["Experiment runners"]
  node_baseline["Baseline trainer<br/>[baseline_main.py]"]
  node_federated["Federated coordinator<br/>[federated_main.py]"]
  node_options["Run options<br/>[options.py]"]
end

subgraph group_data["Data preparation"]
  node_dataset_loader["Dataset loader<br/>[utils.py]"]
  node_partitioner["Client partitioning<br/>[sampling.py]"]
  node_dataset_files[("Dataset files")]
end

subgraph group_learning["Models and learning"]
  node_model_architectures["MLP and CNNs<br/>[models.py]"]
  node_local_update["Local training<br/>[update.py]"]
  node_weight_averaging["Weight averaging<br/>[utils.py]"]
  node_evaluation["Model evaluation<br/>[update.py]"]
end

subgraph group_artifacts["Experiment outputs"]
  node_saved_results[("Saved results")]
end

node_user(("Researcher"))
node_torchvision["Torchvision datasets"]
node_tensorboard["TensorBoard logging"]

node_user -->|"runs baseline"| node_baseline
node_user -->|"runs federated"| node_federated
node_baseline -->|"parses options"| node_options
node_federated -->|"parses options"| node_options
node_baseline -->|"loads data"| node_dataset_loader
node_federated -->|"loads data"| node_dataset_loader
node_dataset_loader -->|"loads datasets"| node_torchvision
node_torchvision -->|"downloads or reads"| node_dataset_files
node_dataset_loader -->|"assigns client data"| node_partitioner
node_baseline -->|"constructs model"| node_model_architectures
node_federated -->|"constructs model"| node_model_architectures
node_baseline -->|"tests model"| node_evaluation
node_federated -->|"trains selected clients"| node_local_update
node_local_update -->|"runs model"| node_model_architectures
node_federated -->|"averages client weights"| node_weight_averaging
node_federated -->|"tests global model"| node_evaluation
node_local_update -->|"logs loss"| node_tensorboard
node_baseline -->|"saves plot"| node_saved_results
node_federated -->|"saves metrics"| node_saved_results

click node_baseline "https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/baseline_main.py"
click node_federated "https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/federated_main.py"
click node_options "https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/options.py"
click node_dataset_loader "https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/utils.py"
click node_partitioner "https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/sampling.py"
click node_dataset_files "https://github.com/ashwinrj/federated-learning-pytorch/tree/master/data"
click node_model_architectures "https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/models.py"
click node_local_update "https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/update.py"
click node_weight_averaging "https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/utils.py"
click node_evaluation "https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/update.py"
click node_saved_results "https://github.com/ashwinrj/federated-learning-pytorch/tree/master/save"

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_baseline,node_federated,node_options toneBlue
class node_dataset_loader,node_partitioner,node_dataset_files toneAmber
class node_model_architectures,node_local_update,node_weight_averaging,node_evaluation toneMint
class node_saved_results toneRose
class node_user,node_torchvision,node_tensorboard toneIndigo
```

## Components

### Experiment runners

- [Baseline trainer](https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/baseline_main.py): `src/baseline_main.py`
- [Federated coordinator](https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/federated_main.py): `src/federated_main.py`
- [Run options](https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/options.py): `src/options.py`

### Data preparation

- [Dataset loader](https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/utils.py): `src/utils.py`
- [Client partitioning](https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/sampling.py): `src/sampling.py`
- [Dataset files](https://github.com/ashwinrj/federated-learning-pytorch/tree/master/data): `data`

### Models and learning

- [MLP and CNNs](https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/models.py): `src/models.py`
- [Local training](https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/update.py): `src/update.py`
- [Weight averaging](https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/utils.py): `src/utils.py`
- [Model evaluation](https://github.com/ashwinrj/federated-learning-pytorch/blob/master/src/update.py): `src/update.py`

### Experiment outputs

- [Saved results](https://github.com/ashwinrj/federated-learning-pytorch/tree/master/save): `save`

### Other components

- **Researcher**
- **Torchvision datasets**
- **TensorBoard logging**

## Connections

- Researcher → Baseline trainer: runs baseline (evidence: `README.md`)
- Researcher → Federated coordinator: runs federated (evidence: `README.md`)
- Baseline trainer → Run options: parses options (evidence: `src/baseline_main.py`)
- Federated coordinator → Run options: parses options (evidence: `src/federated_main.py`)
- Baseline trainer → Dataset loader: loads data (evidence: `src/baseline_main.py`)
- Federated coordinator → Dataset loader: loads data (evidence: `src/federated_main.py`)
- Dataset loader → Torchvision datasets: loads datasets (evidence: `src/utils.py`)
- Torchvision datasets → Dataset files: downloads or reads
- Dataset loader → Client partitioning: assigns client data (evidence: `src/utils.py`)
- Baseline trainer → MLP and CNNs: constructs model (evidence: `src/baseline_main.py`)
- Federated coordinator → MLP and CNNs: constructs model (evidence: `src/federated_main.py`)
- Baseline trainer → Model evaluation: tests model (evidence: `src/baseline_main.py`)
- Federated coordinator → Local training: trains selected clients (evidence: `src/federated_main.py`)
- Local training → MLP and CNNs: runs model (evidence: `src/update.py`)
- Federated coordinator → Weight averaging: averages client weights (evidence: `src/federated_main.py`)
- Federated coordinator → Model evaluation: tests global model (evidence: `src/federated_main.py`)
- Local training → TensorBoard logging: logs loss (evidence: `src/update.py`)
- Baseline trainer → Saved results: saves plot (evidence: `src/baseline_main.py`)
- Federated coordinator → Saved results: saves metrics (evidence: `src/federated_main.py`)
