fastai VS ru-dalle

Jeremy Howard's YouTube Channel - Jeremy maintains the fastai library, which is an excellent package that will help anyone build complicated ML architectures in minimum time. His YouTube Channel has a number of free courses which do an amazing job of covering a variety of ML topics, and he also maintains a very active forum for people studying ML.

To create the AI we will use fastai. This is a python library, which is build on top of pytorch. No worries, you don't need to know how to code python. We will learn how this stuff works along the way :)

You may know fast.ai as a popular deep learning course. There is also a deep learning library with the same name (https://github.com/fastai/fastai) as well as software development tools like nbdev (https://nbdev.fast.ai/).

fast.ai has been offering education and tools for free for over 7 years, and has been approached by many companies asking for help. This program offers an avenue for business to get relevant professional services and support.

The "fastai" course is free, and does a really nice job walking you through building simple neural nets from the ground up:

https://github.com/fastai/fastai

What's going on here is the exact same thing, just much, much larger.

Development NB: https://github.com/fastai/fastai/blob/master/nbs/09_vision.augment.ipynb

There is already a ton of momentum around automating ML workflows. I would suggest you contribute to a preexisting project like, for instance, PyTorch Lightning or fast.ai.

If you are unaware of what fastai is, its official description is:

Also, not a single docstring to document any code in the library - https://github.com/fastai/fastai/blob/master/fastai/vision/learner.py

It’s a custom trained model, built in rudalle https://github.com/ai-forever/ru-dalle

It's a custom model, built from rudalle

ru-dalle has had that feature for quite a while, as it was their first inpainting example notebook:

GitHub repo.

Colab notebook. There are actually 5 models, but I doubt the 12B parameter Kandinsky model is actually available per looking at this code.

This was made with a mini version of DALL-E: ruDALL-E

Specifically, I fine-tuned ru-DALLE using a dataset consisting of ~1000 images of imperial denarii (ranging from Augustus through Maximinus Thrax) coupled with descriptions of each coin grabbed from OCRE. For example, the obverse description of this coin would be "Head of Augustus, bare, right" and the reverse description would be "Round shield, spear-head, and curved sword".

One way to use the version 3 model is to use this official Colab notebook linked to in the ruDALL-E GitHub repo. I recommend making the changes mentioned in this post. If you want to use the older version 2 model with this Colab notebook, change 'Malevich' to 'Malevich_v2' in line "dalle = get_rudalle_model('Malevich', pretrained=True, fp16=True, device=device)" (relevant source code).

# !pip install rudalle==0.0.1rc8 > /dev/null !pip3 install git+https://github.com/sberbank-ai/ru-dalle.git@feature/new_malevich