Fuel your custom models on the Cloud AI Platform
summary
Builds and deploys a TensorFlow regression model on Vertex AI from a Kaggle dataset of late 1970s and early 1980s cars using a custom container and notebook workflow. Training jobs run on the platform (demo job takes about 10–15 minutes) and the deployed REST endpoint returns predictions (example shown as ~16 mpg prediction). Codelab and source code are provided and the demo shows both training and serving end to end.
AI-written summary · about
chapters (5)
related talks
11mBecome a Cloud Machine Learning developer in 5 steps
40mVertex AI: Pipelines for your MLOps workflows
18mBuilding Efficient PyTorch Models on Vertex AI
45mVertex AI: Pipelines for your MLOps workflows
22mTrain Custom Machine Learning Models with No Data Science Expertise
33mFaster model training and experimentation with Vertex AI
26mBuilding Portable ML Pipelines for Development and Production
46mDeploying scalable Machine Learning models with Kubernetes