Skip to content

Tools for time series processing and visualization

License

Notifications You must be signed in to change notification settings

srlindemann/amp

Repository files navigation

How to setup

Documentation

  • Documentation is under documentations/notes and in the directory of the corresponding tools

Setting up a tmux session

  • Create a tmux session for lem + amp set-up
    > cd ~/src
    > ./dev_scripts_p1/tmux_p1.sh part1
  • TODO(gp): It might be dependent on GP's set-up and needs to be generalized, if worth it for general consumption

Development flow

  • We package in Docker containers what is needed to run AM system
    • E.g., command lines and unit tests should always run inside a container
  • We don't development tools save inside container, but we use the local machine to develop (e.g., running PyCharm)
  • Some workflows are run on a development machine without Docker
    • We install a light virtual environment with the minimum set of dependency
      • We use pyinvoke, a replacement for make written in Python, to create workflows
    • invoke always runs outside Docker and in a shell on the dev machine
    • The code for the client setup is under dev_scripts/client_setup

Setting up

  • Build the thin virtual env on the dev machine

    > dev_scripts/client_setup/build.sh
    
  • Activate the virtual env

    > source dev_scripts/client_setup/activate.sh
    
  • Configure the env

    • source dev_scripts/setenv_amp.sh

Docker

  • Create a docker bash to run interactively (e.g., pytest or command lines)

    # Pull the container.
    > invoke docker_pull
    > invoke docker_bash
    
  • Run fast tests:

    > invoke run_fast_tests

About

Tools for time series processing and visualization

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Contributors 20