Serving Multiple TensorFlow Models with Monitoring
Serve multiple TensorFlow models simultaneously and monitor them with Prometheus and Swagger.
Posts
Deploying TensorFlow model as Canary Releases Kubernetes and Istio
Use Istio with Minikube and TensorFlow Serving to create canary deployments of TensorFlow machine learning models!
Prepare and Deploy a TensorFlow Model to Tensorflow Serving
Serve tensorflow model efficiently with customized model signatures
Integrating Seldon Core with Kafka
Stream predictions in and out of Seldon Core model deployments using Kafka.
Building Inference Graphs with Seldon Core
Compose multi-step inference graphs (transformers, combiners, routers) with Seldon Core.
Orchestrating ML Workflows with Seldon Core
Automate model deployment workflows on top of Seldon Core.
Deploying Your First Model with Seldon Core
Deploy a first machine learning model on Kubernetes using Seldon Core, including authentication basics.
Building Kubeflow Pipeline Components
Create custom and reusable components for Kubeflow Pipelines.
Distributed Training with TensorFlow
Train TensorFlow models across multiple workers using distributed training strategies.
Autoscaling TensorFlow Serving on Kubernetes
Set up horizontal pod autoscaling for a TensorFlow Serving deployment on Kubernetes.