An image model created on the base of okrol model by me
Install the requirements pip install -r requirements.txt,
Install the language processer python -m spacy download en_core_web_sm
The model is not really great and shouldn't be used for anything else than experimenting:
- It losses some data from epoches
- It is unstable
- It bugs out very often
- The diffusion algorithm is not very good
This is a blueprint for example data
[
{
"text": "A cat sitting on a couch.",
"image": "images/cat1.jpg"
}
]
#Note in this example images folder will be in the root folder not the traing_data folder Ai was used for generating docstrings/comments