Our great sponsors
-
Recently I've been learning MLOps. There's a lot to learn, as shown by this and this repository listing MLOps references and tools, respectively.
-
However, installing and trying out Kubeflow Pipelines (KFP) is a lot simpler. In this post, we'll create a local cluster with kind, install KFP as described here and run our first pipeline.
-
Scout APM
Less time debugging, more time building. Scout APM allows you to find and fix performance issues with no hassle. Now with error monitoring and external services monitoring, Scout is a developer's best friend when it comes to application development.
-
Recently I've been learning MLOps. There's a lot to learn, as shown by this and this repository listing MLOps references and tools, respectively.
-
Kubeflow has multiple components: central dashboard, Kubeflow Notebooks to manage Jupyter notebooks, Kubeflow Pipelines for building and deploying portable, scalable machine learning (ML) workflows based on Docker containers, KF Serving for model serving (apparently superseded by KServe), Katib for hyperparameter tuning and model search, and training operators such as TFJob for training TF models on Kubernetes.
-
The code for this example can be found in this repository.
Related posts
- [HIRING][Full Time, Part Time, Temporary, Internship, Freelance] Data Science Intern (Remote)
- What do you think about Detecting Silent ML Failure with an Open Source Python library?
- Can I estimate the impact of data drift on performance?
- [D] Tips for ML workflow on raw data
- [P] DagYard - DVC x MLflow x Colab x Gdrive - Automatically Configured