bugs with "initialize_from_state"🐛 [BUG]

[wwwjjj123yyyy]

When I want to statrt a new trainning with different settings from previous trainning model, I change some settings in my .yaml file and add something looks like the followings:

But there is some error ：

I don't know how to handdle it. Is there anybody who can serve some instructions?

[Linux-cpp-lisp]

What version are you using? This looks correct in the versions I'm looking at.

[wwwjjj123yyyy]

Pytorch: 1.11.10
python: 3.9.16
the nequip package is from https://github.com/mir-group/nequip
the allegro exstension package is from https://github.com/Hongyu-yu/allegro

[Linux-cpp-lisp]

Maybe try the develop branch of nequip.

[wwwjjj123yyyy]

Maybe try the develop branch of nequip.

Where is the develop branch of nequip, please?

[ligerzero-ai]

https://github.com/mir-group/nequip/tree/develop

Clone this instead using

git clone https://github.com/mir-group/nequip.git
cd nequip
git checkout develop
pip install .

and install using that folder. Make sure you're in a new environment, or remove nequip in your environment before you do this.