Seldon Core v1 Runtime Demos

The demos assume you have a running Seldon Enterprise Platform installed with relevant permissions to create/update deployments.

They mostly use pre-built models, except the kubeflow demo which has steps to build a model and push it to minio.

Models can be pushed to minio or other objects stores for pre-packaged model servers or packaged as docker containers using language wrappers. See the Seldon Core docs for the building and hosting stages.

However you install your SeldonDeployments, Seldon Enterprise Platform should see them in visible namespaces. These demos use the Enterprise Platform UI for ease.

Demos List

Feature

Demo

Model Explanation

- Tabular Data Classification on Income Dataset
- Image Classification on CIFAR10 Dataset
- Text Sentiment on IMDB Movie reviews Dataset

Model and Data Monitoring

- Feature Distributions
- Drift Detection
- Outlier Detection
- Accuracy Metrics

Deployment Servers and
Rollout Strategies

- Canary Promotion
- Text Generation with Custom HuggingFace Model
- NVIDIA Triton Server

Batch Jobs

- Prediction Jobs

Security and System Monitoring

- Project Level Auth