Skip to content

Lume-epics is a dedicated API for serving LUME model variables with EPICS.

License

Notifications You must be signed in to change notification settings

kathryn-baker/lume-epics

 
 

Repository files navigation

lume-epics

Lume-epics is a dedicated API for serving LUME model variables with EPICS. Configurations for LUME model variables can be found in lume-model.

Installation

Lume-epics may be installed via conda on the conda-forge channel:
$ conda install lume-epics -c conda-forge

Alternatively, you may install from the GitHub repository using:
$ pip install https://github.com/slaclab/lume-epics.git

Server

The EPICS server requires a model class, model_kwargs, and an epics configuration for instantiation. Once instantiated, the server is started using the Server.start() method, which has an optional monitor keyword argument, monitor, that controls thread execution. When monitor=True, the server is run in the main thread and may be stopped using keyboard interrupt (Ctr+C). If using monitor=False, the server can be stopped manually using the Server.stop() method.

from lume_epics.epics_server import Server
from lume_model.utils import variables_from_yaml
from lume_epics.utils import config_from_yaml
import os
from examples.model import DemoModel

import logging


if __name__ == "__main__":
    with open("examples/files/demo_config.yml", "r") as f:
        input_variables, output_variables = variables_from_yaml(f)

    with open("examples/files/epics_config.yml", "r") as f:
        epics_config = config_from_yaml(f)

    server = Server(
        DemoModel,
        epics_config,
        model_kwargs={
            "input_variables": input_variables,
            "output_variables": output_variables,
        },
    )
    # monitor = False does not loop in main thread
    server.start(monitor=True)

Compatable models

See docs for notes on serving online models with lume-epics.

About

Lume-epics is a dedicated API for serving LUME model variables with EPICS.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%