Make training faster by removing explicit checking for parameters not receiving gradients

Format code with black and isort :)

Remove deep copying as it causes problems with accelerate

Update README.md

Create FUNDING.yml

Merge pull request #2 from AntreasAntoniou/add-code-of-conduct-1

Create CODE_OF_CONDUCT.md

Update issue templates

Merge branch 'main' of https://github.com/AntreasAntoniou/minimal-ml-template

Format code, and update dockerfile

Format code with black and isort :)

Update format-code.yml

Update format-code.yml

Update format-code.yml

Update format-code.yml