Basic GUI to use OpenAI

Motivation

Some time ago, I did some testing of image variations with OpenAPI with a ink sketch I did some years ago (is quite easy to identify the original).

I showed the generated image to my brother and ask him to find which image is the original.

Recently, my brother needed to generate some image variations for their projects, and as he don't know Python, programming or API stuffs, he asked me if I could make them for him, and I did it.

Then, I made this basic GUI for this task.

This program/GUI is pretty basic, but functional

Some references

• OpenAI Image variation
• OpenAI Image variation - API

Usage

1. Create an account in OpenAI
2. Get an Api Key https://platform.openai.com/docs/api-reference/authentication
3. Open program
4. Fill "OpenAI Api Key" with your api key (after a succesful execution will be saved to "config.json")
5. Select an image (Select image)
6. Choose number of variation to generate (Num of variation)
7. Choose ouput image size (Image Size)
8. Optional add a Image prefix
9. Presss Execute

Variation images will be copied to folder variations as <prefix><image filename><remote_filename>.png

About the program

If you want to build the program:

• Install Python 3.x

• Create a virtualenv, activate and install the requirements

  python -m venv venv
  source venv/bin/activate
  python -m pip install --upgrade pip
  pip install -r requirements.txt
  

  (*) For Windows users use venv\Scripts\activate

• Build binary, executing:

  • Windows

    pyinstaller --clean --onefile  --noconsole --windowed --name "OpenAI_GUI"  --icon=images/logo.ico  --add-data="images/*;images/" gui_openai.py
    

  • Linux

  pyinstaller --clean --onefile  --noconsole --windowed --name "OpenAI_GUI"  --icon=images/logo.ico  --add-data="images/*:images/" gui_openai.py
  

Check binary in dist folder