Skip to content

Repository files navigation

Nice-ODE Stack

Introduction

  • This repository contains the instructions to build the containerized Nice-ODE stack from a linux machine
    • That is, linux desktop/cli -> linux container. Windows -> linux container is not supported at this time.
  • Nice-ODE is a framework for optimizing the parameters of ODE's.
  • Before building the stack you will need to install Docker.

Components

  • ./dockerfile
    • The central development container.
    • Installs dependencies needed to work with an 'advanced' Python setup
      • CUDA
      • Pytensor (PyMC dependency)
      • R integration
      • Poetry managed python dev-tools
  • ./mlflow_image/
    • A local mlflow (https://mlflow.org/) server to track your work
    • Connects to a default PostgreSQL container defined in Nice-ODE's runtime docker compose file (not in this repo)
  • ./gemma_llm/
    • Not constructed by default
    • A local Gemma 3 sever
    • You will need you own Hugging Face API key provisioned with read access to download the model weights.
    • Does not provide any additional functionality to the Nice-ODE stack.

Building

  1. Clone this repo to your PC.

  2. Open Terminal and cd into the directory where the repo was cloned

    1. eg. git_repos/nice-ode-stack
  3. Run the bash script docker-compose-timestamp.sh as shown below.

    1. The compose log will be stored in ./docker_compose.log.
    2. If you run the command multiple times, consider updating the command to save the log to, for example, docker_compose2.log, docker_compose3.log etc.
  • See the docker_compose.log_example if you want to check out what your log should look like.
  • docker_compose.log_example is a text file and can be opened in any text editor.
  • The file type is .log_example rather than .log because .log is ignored by git.
your-machine-name:~/git_repos/nice-ode-stack$ ./docker-compose-timestamp.sh > docker_compose.log 2>&1

About

Containerized Stack for Nice-ODE

Topics

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages