Seamlessly Orchestrate ML Pipelines on Azure with ZenML
Integrate the power of Microsoft Azure with ZenML to effortlessly orchestrate and manage your machine learning pipelines in the cloud. This integration allows you to leverage Azure's scalable infrastructure and comprehensive ML services while benefiting from ZenML's streamlined workflow management capabilities.
Features with ZenML
- Seamless deployment of ZenML pipelines on Azure infrastructure
- Scalable compute resources for efficient model training and inference
- Enable collaboration by sharing artifacts across teams and stakeholders
- Flexible orchestrator settings for serverless, compute instance or compute cluster modes
- Secure access to artifacts using Azure authentication methods
Main Features
- Comprehensive cloud-based environment for the entire ML lifecycle
- Scalable compute resources and managed services for ML workloads
- Visual interface for monitoring and managing ML experiments and models
- Seamless integration with other Azure services and tools
- Enterprise-grade security and compliance features
How to use ZenML with
Microsoft Azure
# Register a ZenML Azure stack by using existing infrastructure
# zenml stack register <STACK_NAME> -p azure
# OR, create a ZenML Azure stack by deploying new infrastructure
# zenml stack deploy -p azure
from zenml import pipeline, step
@step
def hello_world() -> str:
return "Hello World!"
@pipeline
def my_pipeline():
_ = hello_world()
if __name__ == "__main__":
my_pipeline()
The code example demonstrates how to register an Azure stack with ZenML:
- Use the CLI to register a stack using either by using existing infrastructure or by deploying it anew
- Use the
@step and @pipeline decorators to define a pipeline and run it on your new Azure stack.
Additional Resources
Set up a minimal Azure stack
Read a detailed Azure guide on how to establish a ZenML Azure stack and execute your pipelines
Deploy an Azure stack using the 1-click deployment tool
Register an Azure stack using the stack wizard
.png)
