Picture creator

Creates a picture using stable diffusion 1.5
Everything is configurable within the main file and environment variables
After running once you can run always offline, the first time is to download the model
The examples bellow were made with CPU i5-1035G1 & 6GB of RAM

Examples made with this code:

Existing editing features:

Normal	Enhanced	Grey scaled	Negative	Hide specific color

What you need to install to run this:

pip install diffusers transformers accelerate pillow

How to run:

python main.py

Where all models installed:

~/.cache/huggingface/hub

TODO:

1. Add validation for prompt to initiate specific agents after creation of the image
2. Add text generation model that will improve the given prompt and negative prompt
3. Add upscale model to boost image visualization
4. Decide if input will be reading from Kafka or GRPC and implement
5. Tests
6. Docker