# ModuleStudy-CIFAR10

**Repository Path**: maotao_break/module-study-cifar10

## Basic Information

- **Project Name**: ModuleStudy-CIFAR10
- **Description**: 基于LeNet、AlexNet和MobileNet 架构训练CIFAR-10数据集
- **Primary Language**: Python
- **License**: MPL-2.0
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2025-05-09
- **Last Updated**: 2025-05-25

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# 基于LeNet、AlexNet和MobileNet 架构训练CIFAR-10数据集
<!-- TOC -->
* [基于LeNet、AlexNet和MobileNet 架构训练CIFAR-10数据集](#基于lenetalexnet和mobilenet-架构训练cifar-10数据集)
  * [训练集](#训练集)
  * [训练](#训练)
  * [结果](#结果)
    * [模型 weights/](#模型-weights)
    * [曲线 training_plots/](#曲线-training_plots)
  * [应用](#应用)
<!-- TOC -->
## 训练集

1. 下载
   [https://www.cs.toronto.edu/~kriz/cifar.html](https://www.cs.toronto.edu/~kriz/cifar.html)
2. 解压并放入目录内：cifar-10-batches-py/

## 训练

1. 选择模型：在 trainer.py 内的 main 中

```python

if __name__ == '__main__':
    # 示例：训练MobileNet、LeNet
    for model in [MobileNet, LeNet]:
        trained_model = train_model(model(), epochs=20)

        torch.save(trained_model.state_dict(), f"weights/cifar10_{model.__name__}.pth")

```

2. 运行[trainer.py](trainer.py)

## 结果

### 模型 weights/

* cifar10_AlexNet.pth
* cifar10_LeNet.pth
* cifar10_MobileNetCIFAR.pth

### 曲线 training_plots/

![](doc/AlexNet_training_curves.png)
![](doc/MobileNetCIFAR_training_curves.png)
![](doc/LeNet_training_curves.png)

## 应用

运行[test.py](test/test.py)