deepimg3 1.4.1

Automated testing for the DeepAI text to image api

deepimg3

deepimg3 is a Python module designed for automated testing of the DeepAI Text-to-Image API. It allows users to asynchronously and efficiently generate batches of images based on specified sets of prompts, managing the process through command-line arguments. The tool supports selecting image quality levels and managing multiple prompts through JSON configuration files.

Prerequisites

• Python 3.12 or higher
• httpx
• tqdm

Installation

Download the latest .whl file from the project repository and install it using pip:

pip install deepimg3

Usage

After installation, you can run the module from the command line:

python -m deepimg3 --help

This will display help information about all available command-line options:

Usage: __main__.py [-h] [-n NUM] [-p PROMPTS] [-q {both,hd,sd}] [-c CONCURRENCY]

Generate images based on command line arguments.

options:
  -h, --help            show this help message and exit
  -n NUM, --num NUM     Number of images to generate for each prompt/quality
  -p PROMPTS, --prompts PROMPTS
                        File path to the promptsets file. (JSON)
  -q {both,hd,sd}, --quality {both,hd,sd}
                        Quality of the image to generate.
  -c CONCURRENCY, --concurrency CONCURRENCY
                        Number of concurrent image generation tasks to run. Default: 4

Examples

Run the tool to generate images using the default promptsets.json:

python -m deepimg3 --prompts promptset.example.json --num 2 --quality hd

This command will generate 2 HD images per prompt found in promptsets.example.json.

Configuration

See promptsets.example.json for an example of how to format this file.

Development

If you want to contribute to deepimg3, you can clone the repository and install the dependencies as follows:

git clone https://github.com/DJStompZone/deepimg3.git
cd deepimg3
poetry install

Make sure you have Poetry installed to manage dependencies.

License

This project is licensed under the MIT License - see the LICENSE file for details.