An intuitive, easy-to-use python interface for batch resource requesting, access, job submission, and observation. Simplifying the developer's life while enabling access to high-performance compute resources, either in the cloud or on-prem.
For guided demos and basics walkthroughs, check out the following links:
- Guided demo notebooks available here, and copies of the notebooks with expected output also available
- these demos can be copied into your current working directory when using the
codeflare-sdkby using thecodeflare_sdk.copy_demo_nbs()function - Additionally, we have a video walkthrough of these basic demos from June, 2023
Full documentation can be found here
Can be installed via pip: pip install codeflare-sdk
CodeFlare SDK uses kube-authkit for Kubernetes authentication, supporting multiple authentication methods:
- Auto-Detection - Automatically detects kubeconfig or in-cluster authentication
- Token-Based - Authenticate with API server token
- OIDC - OpenID Connect authentication with device flow or client credentials
- OpenShift OAuth - Native OpenShift OAuth support
- Kubeconfig - Traditional kubeconfig file authentication
- In-Cluster - Service account authentication when running in a pod
from kube_authkit import get_k8s_client, AuthConfig
from codeflare_sdk import set_api_client, Cluster, ClusterConfiguration
# Option 1: Auto-detect authentication (recommended - no explicit auth needed!)
cluster = Cluster(ClusterConfiguration(
name='my-cluster',
num_workers=2,
))
cluster.apply()
# Option 2: OIDC authentication
auth_config = AuthConfig(
method="oidc",
oidc_issuer="https://your-oidc-provider.com",
client_id="your-client-id",
use_device_flow=True
)
api_client = get_k8s_client(config=auth_config)
set_api_client(api_client) # Register with CodeFlare SDK
# Option 3: OpenShift OAuth with token
auth_config = AuthConfig(
k8s_api_host="https://api.example.com:6443",
token="your-token"
)
api_client = get_k8s_client(config=auth_config)
set_api_client(api_client) # Register with CodeFlare SDK
# Now create your cluster
cluster = Cluster(ClusterConfiguration(
name='my-cluster',
num_workers=2,
))
cluster.apply()Please see our CONTRIBUTING.md for detailed instructions.
It is possible to use the Release Github workflow to do the release. This is generally the process we follow for releases
The following instructions apply when doing release manually. This may be required in instances where the automation is failing.
- Check and update the version in "pyproject.toml" file.
- Commit all the changes to the repository.
- Create Github release (https://docs.github.com/en/repositories/releasing-projects-on-github/managing-releases-in-a-repository#creating-a-release).
- Build the Python package.
poetry build - If not present already, add the API token to Poetry.
poetry config pypi-token.pypi API_TOKEN - Publish the Python package.
poetry publish - Trigger the Publish Documentation workflow