Difference between revisions of "Note Kubeflow"
From ESS-WIKI
Line 9: | Line 9: | ||
#To speed up your training with GPU, [[K8S_Nvidia_Device_Plugin|Add Nvidia device plugin]] | #To speed up your training with GPU, [[K8S_Nvidia_Device_Plugin|Add Nvidia device plugin]] | ||
#To speed up your training with cluster, [[Distributed_Tensorflow_in_Kubernetes|Distributed Tensorflow in Kubernetes]] | #To speed up your training with cluster, [[Distributed_Tensorflow_in_Kubernetes|Distributed Tensorflow in Kubernetes]] | ||
− | #To export your training model for prediction or inference (TensorFlow-Serving), [[ | + | #To export your training model for prediction or inference (TensorFlow-Serving), [[K8S_TF_Serving_with_yaml|K8S TF Serving with yaml]] |
− | #To export your training model for prediction or inference (Ksonnet), [[ | + | #To export your training model for prediction or inference (Kubeflow/Ksonnet), [[K8S_TF_Serving_with_ksonnet|K8S TF Serving with ksonnet]] |
= Reference = | = Reference = |
Latest revision as of 07:36, 13 December 2018
Introduce
Kubeflow is a set of pods that deploy machine learning toolkit in Kubernetes. It help you to make ML training/serving easiler in Kubernetes.
Topics
- To enable Kubeflow, Install kubeflow with ksonnet
- To make a training with web UI editor, Jupyter Notebooks in Kubernetes
- To speed up your training with GPU, Add Nvidia device plugin
- To speed up your training with cluster, Distributed Tensorflow in Kubernetes
- To export your training model for prediction or inference (TensorFlow-Serving), K8S TF Serving with yaml
- To export your training model for prediction or inference (Kubeflow/Ksonnet), K8S TF Serving with ksonnet