# yolo-inference

**Repository Path**: wangjq4214/yolo-inference

## Basic Information

- **Project Name**: yolo-inference
- **Description**: yolo-inference
- **Primary Language**: Unknown
- **License**: Apache-2.0
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2025-02-18
- **Last Updated**: 2025-02-23

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Yolo Example

## 1. Inference

This is a simple example of how to use Yolo Inference.

### 1.1. How to run the app

We strongly recommend using [Docker](https://docs.docker.com/get-started/) to run this app locally.

#### 1.1.1. Build the Docker image

First, make sure you have [Docker](https://docs.docker.com/get-started/) installed on your machine. And run the following command to build the Docker image:

```bash
docker build -f docker/inference . -t yolo-inference
```

#### 1.1.2. Run the Docker container

To run the Docker container, run the following command:

```bash
docker run -p 8000:8000 yolo-inference
```

### 1.2. The running environment variables

Must specific environment variables:
- `YOLO_INFERENCE_WEIGHT_PATH`: The path to the model file.


Optional environment variables:
- `YOLO_INFERENCE_SAVE_RESULT`: Whether to save the result. (Default: `False`)
- `YOLO_INFERENCE_SAVE_PATH`: The path to save the result. (Default: `./result`)

If you enable `YOLO_INFERENCE_SAVE_RESULT`, the inference result will be saved to `YOLO_INFERENCE_SAVE_PATH`.

## 2. Training

This is a simple example of how to use Yolo Training.

### 1.1. How to run the app

We strongly recommend using [Docker](https://docs.docker.com/get-started/) to run this app locally.

#### 1.1.1. Build the Docker image

First, make sure you have [Docker](https://docs.docker.com/get-started/) installed on your machine. And run the following command to build the Docker image:

```bash
docker build -f docker/train . -t yolo-train
```

#### 1.1.2. Run the Docker container

To run the Docker container, run the following command:

```bash
docker run -p 8000:8000 yolo-train
```

### 1.2. The running environment variables

Must specific environment variables:
- `YOLO_TRAIN_WEIGHT_PATH`: The path to the model file.
- `YOLO_TRAIN_DATASET_PATH`: The path to the dataset file.

Optional environment variables:
- `YOLO_TRAIN_EPOCHS`: The number of epochs to train. (Default: `10`)
- `YOLO_TRAIN_BATCH_SIZE`: The batch size to use during training. (Default: `16`)
- `YOLO_TRAIN_LEARNING_RATE`: The learning rate to use during training. (Default: `0.001`)


### 1.3. The dataset struct

The dataset struct is as follows:

```
.
|- data.yaml
|- images
|--- train
|----- 0.jpg
|--- val
|----- 1.jpg
|- labels
|--- train
|----- 0.txt
|--- val
|----- 1.txt
```