https://github.com/songquanpeng/pytorch-deployment

A template for rapid deployment of PyTorch models.
https://github.com/songquanpeng/pytorch-deployment

flask-application pytorch-template stargan-v2

Last synced: about 1 year ago
JSON representation

A template for rapid deployment of PyTorch models.

• Host: GitHub
• URL: https://github.com/songquanpeng/pytorch-deployment
• Owner: songquanpeng
• Created: 2021-08-20T01:53:57.000Z (almost 5 years ago)
• Default Branch: main
• Last Pushed: 2022-10-08T13:35:48.000Z (over 3 years ago)
• Last Synced: 2025-03-28T17:11:10.579Z (about 1 year ago)
• Topics: flask-application, pytorch-template, stargan-v2
• Language: Python
• Homepage:
• Size: 22.5 KB
• Stars: 64
• Watchers: 2
• Forks: 10
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# PyTorch Deployment

> A template for rapid deployment of PyTorch models.

点击以展开中文 README



## 功能

+ [x] API.

+ [x] Web 前端.

+ [x] 支持图像转换模型.

+ [ ] 支持图像分类模型.

+ [ ] 支持图像生成模型.

## 演示

![starganv2_afhq](https://user-images.githubusercontent.com/39998050/155641683-fbef7d4a-7a44-4f60-bf96-7df79a02c0ee.gif)

## 部署步骤

1. 安装依赖：`pip install -r requirements.txt`

2. 启动服务，你有以下选择：

    1. `flask run -p 3000`，注意该方式性能较差。

    2. 使用 Gunicorn：

        1. 安装 Gunicorn：`pip install gunicorn`

        2. 启动应用：`gunicorn -b 127.0.0.1:3009 app:app`

请注意，模型的权重是从 Github 上下载的，如果你的服务器无法正常访问 Github，请手动下载权重并放到 data 文件夹下。

目前内置的模型：

+ starganv2_afhq.ckpt：https://github.com/songquanpeng/pytorch-deployment/releases/download/v0.1.0/starganv2_afhq.ckpt



## Features

+ [x] API.

+ [x] Web frontend.

+ [x] Support image translation models.

+ [ ] Support image classification models.

+ [ ] Support image generation models.

## Demo

![starganv2_afhq](https://user-images.githubusercontent.com/39998050/155641683-fbef7d4a-7a44-4f60-bf96-7df79a02c0ee.gif)