# wongkinyiu/yolov7 architecture

> Architecture diagram of the GitHub repository wongkinyiu/yolov7, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-09-24). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/wongkinyiu/yolov7
- Repository: https://github.com/wongkinyiu/yolov7

## Overview

YOLOv7 is an object-detection framework for training, evaluating, exporting, and running model variants on images and video. The main workflow prepares labeled data, trains a configured detector, then uses its weights for inference or validation; export and Triton serving provide deployment routes. The graph also shows loss computation and model building as internal layers. README claims guide capabilities and workflow; because several entry-point sources were not sampled, some links rely on documented usage rather than inspected call sites.

Main entry points: [`train.py`](https://github.com/wongkinyiu/yolov7/blob/main/train.py) (Train detector), [`train_aux.py`](https://github.com/wongkinyiu/yolov7/blob/main/train_aux.py) (Auxiliary training), [`models/yolo.py`](https://github.com/wongkinyiu/yolov7/blob/main/models/yolo.py) (YOLO model), [`test.py`](https://github.com/wongkinyiu/yolov7/blob/main/test.py) (Evaluate models)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_training["Training"]
  node_train["Train detector<br/>[train.py]"]
  node_auxtrain["Auxiliary training<br/>[train_aux.py]"]
  node_configs["Training configs"]
  node_data_config["Dataset config<br/>[coco.yaml]"]
  node_loader["Dataset loading<br/>[datasets.py]"]
  node_loss["Detection losses<br/>[loss.py]"]
end

subgraph group_detection["Detector"]
  node_model["YOLO model<br/>[yolo.py]"]
  node_blocks["Model layers<br/>[common.py]"]
end

subgraph group_inference["Inference"]
  node_detect["Run detection<br/>[detect.py]"]
  node_postprocess["Detection utilities<br/>[general.py]"]
end

subgraph group_deployment["Deployment"]
  node_export["Export models<br/>[export.py]"]
  node_reparam["Reparameterize"]
  node_triton_client["Triton client<br/>[client.py]"]
end

subgraph group_validation["Validation"]
  node_test["Evaluate models<br/>[test.py]"]
  node_metrics["Detection metrics<br/>[metrics.py]"]
end

node_user(("Engineer"))
node_dataset[("COCO data")]
node_triton_server{{"Triton server"}}

node_user -->|"starts training"| node_train
node_user -->|"starts training"| node_auxtrain
node_user -->|"provides input"| node_detect
node_user -->|"requests evaluation"| node_test
node_user -->|"requests export"| node_export
node_user -->|"runs procedure"| node_reparam
node_train -->|"reads config"| node_data_config
node_auxtrain -->|"reads config"| node_data_config
node_train -->|"reads model config"| node_configs
node_auxtrain -->|"reads model config"| node_configs
node_train -->|"loads batches"| node_loader
node_auxtrain -->|"loads batches"| node_loader
node_loader -->|"reads images"| node_dataset
node_train -->|"constructs model"| node_model
node_auxtrain -->|"constructs model"| node_model
node_train -->|"computes loss"| node_loss
node_auxtrain -->|"computes loss"| node_loss
node_model -->|"uses layers"| node_blocks
node_train -->|"saves weights"| node_model
node_auxtrain -->|"saves weights"| node_model
node_detect -->|"runs weights"| node_model
node_detect -->|"processes predictions"| node_postprocess
node_test -->|"evaluates weights"| node_model
node_test -->|"loads validation data"| node_loader
node_test -->|"measures results"| node_metrics
node_export -->|"exports model"| node_model
node_triton_client -.->|"sends inference"| node_triton_server

click node_train "https://github.com/wongkinyiu/yolov7/blob/main/train.py"
click node_auxtrain "https://github.com/wongkinyiu/yolov7/blob/main/train_aux.py"
click node_configs "https://github.com/wongkinyiu/yolov7/tree/main/cfg/training"
click node_data_config "https://github.com/wongkinyiu/yolov7/blob/main/data/coco.yaml"
click node_loader "https://github.com/wongkinyiu/yolov7/blob/main/utils/datasets.py"
click node_loss "https://github.com/wongkinyiu/yolov7/blob/main/utils/loss.py"
click node_model "https://github.com/wongkinyiu/yolov7/blob/main/models/yolo.py"
click node_blocks "https://github.com/wongkinyiu/yolov7/blob/main/models/common.py"
click node_detect "https://github.com/wongkinyiu/yolov7/blob/main/detect.py"
click node_postprocess "https://github.com/wongkinyiu/yolov7/blob/main/utils/general.py"
click node_test "https://github.com/wongkinyiu/yolov7/blob/main/test.py"
click node_metrics "https://github.com/wongkinyiu/yolov7/blob/main/utils/metrics.py"
click node_export "https://github.com/wongkinyiu/yolov7/blob/main/export.py"
click node_reparam "https://github.com/wongkinyiu/yolov7/blob/main/tools/reparameterization.ipynb"
click node_triton_client "https://github.com/wongkinyiu/yolov7/blob/main/deploy/triton-inference-server/client.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_train,node_auxtrain,node_configs,node_data_config,node_loader,node_loss toneBlue
class node_model,node_blocks,node_dataset toneAmber
class node_detect,node_postprocess,node_triton_server toneMint
class node_export,node_reparam,node_triton_client toneRose
class node_test,node_metrics,node_user toneIndigo
```

## Components

### Training

- [Train detector](https://github.com/wongkinyiu/yolov7/blob/main/train.py): `train.py`
- [Auxiliary training](https://github.com/wongkinyiu/yolov7/blob/main/train_aux.py): `train_aux.py`
- [Training configs](https://github.com/wongkinyiu/yolov7/tree/main/cfg/training): `cfg/training`
- [Dataset config](https://github.com/wongkinyiu/yolov7/blob/main/data/coco.yaml): `data/coco.yaml`
- [Dataset loading](https://github.com/wongkinyiu/yolov7/blob/main/utils/datasets.py): `utils/datasets.py`
- [Detection losses](https://github.com/wongkinyiu/yolov7/blob/main/utils/loss.py): `utils/loss.py`

### Detector

- [YOLO model](https://github.com/wongkinyiu/yolov7/blob/main/models/yolo.py): `models/yolo.py`
- [Model layers](https://github.com/wongkinyiu/yolov7/blob/main/models/common.py): `models/common.py`

### Inference

- [Run detection](https://github.com/wongkinyiu/yolov7/blob/main/detect.py): `detect.py`
- [Detection utilities](https://github.com/wongkinyiu/yolov7/blob/main/utils/general.py): `utils/general.py`

### Deployment

- [Export models](https://github.com/wongkinyiu/yolov7/blob/main/export.py): `export.py`
- [Reparameterize](https://github.com/wongkinyiu/yolov7/blob/main/tools/reparameterization.ipynb): `tools/reparameterization.ipynb`
- [Triton client](https://github.com/wongkinyiu/yolov7/blob/main/deploy/triton-inference-server/client.py): `deploy/triton-inference-server/client.py`

### Validation

- [Evaluate models](https://github.com/wongkinyiu/yolov7/blob/main/test.py): `test.py`
- [Detection metrics](https://github.com/wongkinyiu/yolov7/blob/main/utils/metrics.py): `utils/metrics.py`

### Other components

- **Engineer**
- **COCO data**
- **Triton server**

## Connections

- Engineer → Train detector: starts training
- Engineer → Auxiliary training: starts training
- Engineer → Run detection: provides input
- Engineer → Evaluate models: requests evaluation
- Engineer → Export models: requests export
- Engineer → Reparameterize: runs procedure
- Train detector → Dataset config: reads config
- Auxiliary training → Dataset config: reads config
- Train detector → Training configs: reads model config
- Auxiliary training → Training configs: reads model config
- Train detector → Dataset loading: loads batches
- Auxiliary training → Dataset loading: loads batches
- Dataset loading → COCO data: reads images
- Train detector → YOLO model: constructs model
- Auxiliary training → YOLO model: constructs model
- Train detector → Detection losses: computes loss
- Auxiliary training → Detection losses: computes loss
- YOLO model → Model layers: uses layers
- Train detector → YOLO model: saves weights
- Auxiliary training → YOLO model: saves weights
- Run detection → YOLO model: runs weights
- Run detection → Detection utilities: processes predictions
- Evaluate models → YOLO model: evaluates weights
- Evaluate models → Dataset loading: loads validation data
- Evaluate models → Detection metrics: measures results
- Export models → YOLO model: exports model
- Triton client → Triton server: sends inference
