Did you think bringing your machine learning model to production was the hard part?
What about model drift?
Now MLOps comes into play but how does it work and what are good tools?
What is:
- Continuous integration (CI)
- Continuous deployment (CD)
- Continuous training (CT)
What about model drift?
Now MLOps comes into play but how does it work and what are good tools?
What is:
- Continuous integration (CI)
- Continuous deployment (CD)
- Continuous training (CT)
The full MLOps life cycle
- Data Engineering: Get and clean the data recurring if necessary
- Model Engineering: Model training, evaluation, testing, and packaging
- Model Deployment: integrating the trained model. Model serving, performance monitoring
- Data Engineering: Get and clean the data recurring if necessary
- Model Engineering: Model training, evaluation, testing, and packaging
- Model Deployment: integrating the trained model. Model serving, performance monitoring
Why is MLOps important?
Just because your model is hitting now doesn't mean it will be doing so 6 months from now
Model drift is real!
- Continuous training (CT)
Just because your model is hitting now doesn't mean it will be doing so 6 months from now
Model drift is real!
- Continuous training (CT)
Why is MLOps important?
If your machine learning model is embedded in a complete project, it is probably not alone, we have an orchestra that plays and needs to be managed
- Continuous integration (CI)
If your machine learning model is embedded in a complete project, it is probably not alone, we have an orchestra that plays and needs to be managed
- Continuous integration (CI)
Why is MLOps important?
When the data changes, the machine learning models need to be readjusted and regularly deployed
- Continuous deployment (CD)
When the data changes, the machine learning models need to be readjusted and regularly deployed
- Continuous deployment (CD)
How do you make your life easier? With suitable tools and automation!
DAGsHub is a web platform for data version control and collaboration for data scientists & ml engineers
It’s like GitHub for data science and machine learning
@TheRealDAGsHub
dagshub.com/docs/faq/index…
DAGsHub is a web platform for data version control and collaboration for data scientists & ml engineers
It’s like GitHub for data science and machine learning
@TheRealDAGsHub
dagshub.com/docs/faq/index…
DAGsHub in production
Continuous Delivery in Machine Learning using Jenkins and DVC an Open Source Version Control System for Machine Learning Projects
dagshub.com/blog/cml-with-…
Continuous Delivery in Machine Learning using Jenkins and DVC an Open Source Version Control System for Machine Learning Projects
dagshub.com/blog/cml-with-…
MLflow is an open source platform to manage the ML lifecycle, including experimentation, reproducibility, deployment, and a central model registry.
@MLflow
mlflow.org
@MLflow
mlflow.org
Weights & Biases: Developer tools for machine learning
Build better models faster with experiment tracking, dataset versioning, and model management
@weights_biases
wandb.ai/site
Build better models faster with experiment tracking, dataset versioning, and model management
@weights_biases
wandb.ai/site
Kubeflow the cloud-native platform for machine learning operations - pipelines, training and deployment.
Deploying and managing a complex ML system at scale.
github.com/kubeflow/kubef…
Deploying and managing a complex ML system at scale.
github.com/kubeflow/kubef…
• • •
Missing some Tweet in this thread? You can try to
force a refresh
