Hera is a Python framework for constructing and submitting Argo Workflows.

This is a tool for Machine Learning engineers to experiment,develop and orchestrate their workflows locally using Python and Argo Workflows

To run this project, you will need to have the following installed:

Devbox || Docker Desktop

 git clone https://github.com/J0hn-B/hera_argo.git && cd hera_argo

# Start a new shell to install the project dependencies
devbox shell

# Deploy and configure Argo Workflows on a local k3d cluster, and get the Argo Workflows UI URL
task deploy

# Install the project dependencies

cd $PWD/src/mlops && poetry install

# Run the example workflows
 poetry run python main.py

• Check the src/mlops/main.py file and src/mlops/mlops folder for the example workflows.

# Access the k3d cluster using kubectl from another terminal

cd hera_argo && devbox shell

# Deploy a yaml example workflow using argo cli

argo submit --watch src/mlops_yaml/first.yaml

• Permissions:

  hera_argo/kubernetes/cluster/overlays/k3d/kustomization.yaml
  

  # Resources to apply
  resources:
    ###?  dir contains config files for a service account with custom permissions.    Uncomment to apply/adjust/modify based on your needs.
    #- ../../base
  
    ###? file contains a role binding for the default service account in the argo    namespace with admin permissions. NOT recommended for production use.
    - default_admin.yaml

# Delete the cluster
task delete

# Exit the shell
exit

# Clean up packages installed by devbox
nix-collect-garbage