prokia / bionavi-np Goto Github PK

Hi,

When I was running the model, I met the module not found error(see figure below).
it said the modulemlp_retrosyn can not be found, I checked the file and found the file is in a different folder, so I add the full path of this module to the python fileprepare_utils.py when import the mlp_retrosyn module. What I did is change the file prepare_utils.py line 9 from mlp_retrosyn.mlp_inference import MLPModel to from retro_star.packages.mlp_retrosyn.mlp_retrosyn.mlp_inference import MLPModel This problem was solved.

However, other modules still can not be found e.g.rdchiral (see figure). So I am wondering, instead of adding the full path to the python file mannually to import the missing module, is there a better solution?

Thanks in advance!

Hi, I notice you have a fold named singlestep, do you have code to make single-step inference of retrosynthesis prediction?

Thanks

Hello,

I am trying to install on Ubuntu following the listed instructions under setup. Everything seems to work fine until it gets to setup.py for mlp-retrosyn. After that everything follows the same error code: 1. It says easy_install command is deprecated.

Thank you!

Dear author,

I am trying to run your code. I did all the instruction you provided on Github.

When I run "python run_biosynthesis.py"

The, I get the following error:
FileNotFoundError: [Errno 2] No such file or directory: '../singlestep/checkpoints/np-like/model_step_30000.pt'

Any idea why? Thanks.

How do I retrain a model?

Hi, thanks for this amazing work.
Nevertheless, lately as I would like to try the BioNavi server, it always said

Failed to submit! The server is under maintenance, Please try later.

Could you please help?
Best regards,

Hi,

When I was running the run_biosynthesis.py file, I got this error: pynvml.nvml.NVMLError_LibraryNotFound: NVML Shared Library Not Found and I thought this is because there is only cpu on my computer.

Could you help me figure out this problem and I wonder is there a cpu version to run your code?

Thanks!