lammps to run DPA2 and DPA3

[bawanglong-2]

After a Easy installation, I ran the example in the water directory under lmp in the example folder, using the potential file frozen with DPA2 , and encountered an error as shown in the figure. What could be the issue?

DeePMD-kit Version

DeePMD-kit v3.0.3

Backend and its version

As picture show

Python Version, CUDA Version, GCC Version, LAMMPS Version, etc

No response

Details

I run this leading.

About my input :

[njzjz]

Please run $PREFIX/bin/lmp

or if you have activated the environment, just run lmp.

[bawanglong-2]

Please run $PREFIX/bin/lmp  请运行 $PREFIX/bin/lmp

or if you have activated the environment, just run lmp.或者如果你已经激活了环境，直接运行 lmp 。

Thank you very much for the teacher's explanation. I can run the frozen model - frozen_model.pth of DPA2 through it, but when it comes to the DPA3 model - DPA-3.1-3M.pt (which the official website indicates cannot be frozen), it starts to report this error.

[njzjz]

DPA-3.1-3M.pt (which the official website indicates cannot be frozen)

FYI @iProzd

[bawanglong-2]

DPA-3.1-3M.pt (which the official website indicates cannot be frozen)DPA-3.1-3M.pt（官方网站表示无法冻结）

FYI   供参考@iProzd

Teacher, I found the DPA3 file here. Using the code for its direct freezing model, I noticed that it shows this. Is it because the instruction is incorrect?

[iProzd]

@bawanglong-2 You missed the --model-branch (or --head )option here, just as the instruction shows.

[bawanglong-2]

@bawanglong-2 You missed the --model-branch (or --head )option here, just as the instruction shows.
Teacher, perhaps I didn't fully grasp your meaning. It's still showing an error.

[iProzd]

@bawanglong-2 You missed the --model-branch (or --head )option here, just as the instruction shows.
Teacher, perhaps I didn't fully grasp your meaning. It's still showing an error.

@bawanglong-2 Please make sure you are using dp in v3.1.0 version (run dp --version to check), and it works for me to run the following command:

dp --pt freeze -c DPA-3.1-3M.pt -o dpa3_water_branch.pth --model-branch H2O_H2O_PD

[pillowill]

Hello, I am currently trying to run LAMMPS as well, but the error I encountered is shown below. Could you please provide your installation process and example files for my reference?