Shulou(Shulou.com)11/24 Report--

The first self-developed full open source natural language low code GPT application rapid development framework-PromptAppGPT is coming!

Recently, Zhang Changwang, a member of the CCF theoretical computer Science and Technology Committee, developed a rapid application development framework PromptAppGPT based on low-code cues, which can realize the rapid application development of natural language based on GPT.

PromptAppGPT includes functions such as low code prompt (Prompt) application development, GPT text generation, DALL-E image generation, online code editor + compiler + runner, automatic user interface generation, support for plug-in extensions, and so on.

Project website: https://promptappgpt.wangzhishi.net/

Project Code: https://github.com/ mleoking / PromptAppGPT

PromptAppGPT provides multi-task conditional triggering, result verification, and failure retry capabilities, which can turn manual generation tasks that would otherwise require multiple steps into automatic completion.

At the same time, users no longer need to memorize and enter tedious Prompt spells, but can easily complete the task by entering the necessary information at the core of the task.

1. Project features low code hint (Prompt) rapid application development

Support for GPT3/4 executors for text generation

Support for Dalle actuators for image generation

Support for using executor (plug-in) extensions

Built-in online code editor, compiler and runner

Automatically generate user interface

English and Chinese bilingual user interface

two。 Quick start 2.1. Gets the API key for OpenAI. 2.2. Visit the website PromptAppGPT and select the web startup or download program: https://promptappgpt.wangzhishi.net/

2.3. Set up the OpenAI key / OpenAI Api Proxy / OpenAI GPT model. OpenAI key: the API key obtained from OpenAI.

OpenAI Api Proxy: you can visit the proxy website of OpenAI API. For example, you can use https://api.openai-proxy.com in China, and you'd better have your own proxy website.

OpenAI GPT model: gpt-4 / gpt-3.5-turbo

2.4. Select and run an application.

2.5. Edit and compile the application.

3. App instance 3.1. Master of Tourism Planning App Code:

Automatically generated App interface and running results:

Successfully generate a detailed tourism plan for Hainan

3.2. APP code of intelligent image generator: (includes two steps of generating drawing Prompt using GPT and drawing using Dalle)

Using PromptAppGPT, you can automate a manual image generation task that would otherwise require two steps. At the same time, there is no need for users to enter their own tedious Prompt mantra, users only need to enter the drawing description they care about.

Automatically generated App interface and running results:

App first uses GPT to generate drawing Prompt pictures.

Then App automatically calls Dalle to generate pictures based on the drawing Prompt output by GPT.

The following figure shows the final picture to generate the resulting picture.

GPT helps to expand the original simple description of "a giant panda" in many English details, such as pandas sitting on bamboo leaves, in the forest, pandas with round ears, and so on.

These expanded descriptions make subsequent DALL-E drawings more detailed and more vivid.

3.3. Multilingual translation of APP codes:

Automatically generated App interface and running results:

Successfully translate Chinese into English

4. Application development PromptAppGPT applications are developed in low code based on YAML format.

The following is a sample application code:

-author: Leoname: master of Tourism Planning description: master of Tourism Planning gptRound: singlefailedRetries: 2sysTask:userTask:-executor: gpt prompt: | I want to play [$I {Tourism destination @ input}]. Please, as a professional tour guide, help me make a travel strategy for [$I {Tourism days @ select#1/2/3/4/5/6/7}]. In addition, I hope the whole process is not too tight, I prefer quiet places, you can simply visit. My budget is about [$I {tourism budget yuan): @ select#1000/2000/3000/4000/5000/6000/7000}] yuan. The extra:author part is the name of the author; the name part is the name of the application; the description part is the description of the application; the gptRound part determines whether to use gpt for single round (single) or multiple round (multiple) conversations. For most applications, the value should be the number of retries or invalid output if the single;failedRetries part sets the failure.

The sysTask section is a collection of tasks separated by-that sets the behavior of the gpt. For many applications, this field can be left blank.

The userTask section contains user-defined tasks, separated by -. Each task must define the prompt and executor properties, while the properties of trigger, outputer, and validator are optional. The application iterates through user tasks in an orderly manner, using the output of the previous task to match the trigger of each task, and the first task to match through trigger is the currently running task.

The author introduces Zhang Changwang, senior researcher and member of CCF theoretical computer Science and Technology Committee. He was a member of the Advanced programming Committee (SPC) of the 37th AAAI Conference on artificial Intelligence (AAAI-23).

He received his master's and doctorate degrees from University College London (UCL) in 2011 and 2015, respectively. He worked in LBS data mining in Alibaba from 2016 to 2017 and advertising recommendations and user portraits in Tencent from 2018 to 2022.

The current research direction is the research and application of information retrieval (search and promotion), natural language processing and big data mining.

Reference:

Https://github.com/mleoking/PromptAppGPT

This article comes from the official account of Wechat: Xin Zhiyuan (ID:AI_era)

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.