Get Started with Azure Machine Learning
summary
Shows end-to-end steps using MNIST to train and deploy with Azure Machine Learning: create a workspace, upload data to Blob, provision a GPU cluster that autos-scales min 4 max 20 nodes, submit jobs with a TensorFlow estimator, track runs in Jupyter or VS Code, run hyperparameter search with HyperDrive, and use deploy to Azure Container Instance via a scoring script (init/run), Docker image and YAML of dependencies. Claims also include AutoML, PyTorch support, Nvidia GPUs and FPGA scoring.
AI-written summary · about
chapters (6)
- 0:00Introduction to Azure ML
- 1:03End-to-end ML workflows
- 2:17Training custom models
- 5:08Optimizing with Hyperdrive
- 6:23Automated ML features
- 7:19Deployment and orchestration
related talks
48mStreamlining your Machine Learning projects with Azure Machine Learning
43mA lap around Azure Machine Learning services
42mPython SDK to train, package and deploy models
44mGetting started with Azure Machine Learning services
58mDeep Dive into Azure Automated Machine Learning
51mDeveloping Solutions with Azure Machine learning studio
25mFrom Data Exploration to Modeling to Deployment
01h 01mAI and ML imbedded in Azure Data Platform