Skip to content

Commit

Permalink
Sync kserve manifests for v0.8.0 (kubeflow#2240)
Browse files Browse the repository at this point in the history
* Update kserve manifests from v0.8.0

Update kserve/kserve manifests from 8079f375cbcedc4d45a1b4aade2e2308ea6f9ae8

Update kserve/models-web-app manifests from v0.8.0

* add auto script for kserve and kserve models webapp manifests

* update readme and kustomization

* add rretry in kserve install for webhook

* ignore error while installing

* move cmds to script to allow local testing and remove sleep

* move crd wait outside set +e

(cherry picked from commit c3db49b)
  • Loading branch information
surajkota authored and VaishnaviHire committed Aug 16, 2022
1 parent 389c29d commit a66e7d1
Show file tree
Hide file tree
Showing 14 changed files with 14,422 additions and 8,714 deletions.
7 changes: 1 addition & 6 deletions .github/workflows/kserve_kind_test.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -27,9 +27,4 @@ jobs:
run: ./tests/gh-actions/install_cert_manager.sh

- name: Build & Apply manifests
run: |
cd contrib/kserve
kubectl create ns kubeflow
kustomize build kserve | kubectl apply -f -
kustomize build models-web-app/overlays/kubeflow | kubectl apply -f -
kubectl wait --for=condition=Ready pods --all --all-namespaces --timeout 180s
run: ./tests/gh-actions/install_kserve.sh
3 changes: 2 additions & 1 deletion README.md
Original file line number Diff line number Diff line change
Expand Up @@ -51,7 +51,8 @@ This repo periodically syncs all official Kubeflow components from their respect
| Tensorboards Web App | apps/tensorboard/tensorboards-web-app/upstream | [v1.5.0](https://github.com/kubeflow/kubeflow/tree/v1.5.0/components/crud-web-apps/tensorboards/manifests) |
| Volumes Web App | apps/volumes-web-app/upstream | [v1.5.0](https://github.com/kubeflow/kubeflow/tree/v1.5.0/components/crud-web-apps/volumes/manifests) |
| Katib | apps/katib/upstream | [v0.14.0-rc.0](https://github.com/kubeflow/katib/tree/v0.14.0-rc.0/manifests/v1beta1) |
| KServe | contrib/kserve/upstream | [v0.7.0](https://github.com/kserve/kserve/tree/v0.7.0) |
| KServe | contrib/kserve/kserve | [release-0.8](https://github.com/kserve/kserve/tree/8079f375cbcedc4d45a1b4aade2e2308ea6f9ae8/install/v0.8.0) |
| KServe Models Web App | contrib/kserve/models-web-app | [v0.8.0](https://github.com/kserve/models-web-app/tree/v0.8.0/config) |
| Kubeflow Pipelines | apps/pipeline/upstream | [1.8.2](https://github.com/kubeflow/pipelines/tree/1.8.2/manifests/kustomize) |
| Kubeflow Tekton Pipelines | apps/kfp-tekton/upstream | [v1.2.1](https://github.com/kubeflow/kfp-tekton/tree/v1.2.1/manifests/kustomize) |

Expand Down
256 changes: 256 additions & 0 deletions contrib/kserve/kserve/kserve-runtimes.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,256 @@
apiVersion: serving.kserve.io/v1alpha1
kind: ClusterServingRuntime
metadata:
name: kserve-lgbserver
spec:
containers:
- args:
- --model_name={{.Name}}
- --model_dir=/mnt/models
- --http_port=8080
- --nthread=1
image: kserve/lgbserver:v0.8.0
name: kserve-container
resources:
limits:
cpu: "1"
memory: 2Gi
requests:
cpu: "1"
memory: 2Gi
supportedModelFormats:
- autoSelect: true
name: lightgbm
version: "2"
---
apiVersion: serving.kserve.io/v1alpha1
kind: ClusterServingRuntime
metadata:
name: kserve-mlserver
spec:
containers:
- env:
- name: MLSERVER_MODEL_IMPLEMENTATION
value: '{{.Labels.modelClass}}'
- name: MLSERVER_HTTP_PORT
value: "8080"
- name: MLSERVER_GRPC_PORT
value: "9000"
- name: MODELS_DIR
value: /mnt/models
image: docker.io/seldonio/mlserver:0.5.3
name: kserve-container
resources:
limits:
cpu: "1"
memory: 2Gi
requests:
cpu: "1"
memory: 2Gi
supportedModelFormats:
- name: sklearn
version: "0"
- name: xgboost
version: "1"
- name: lightgbm
version: "3"
- autoSelect: true
name: mlflow
version: "1"
---
apiVersion: serving.kserve.io/v1alpha1
kind: ClusterServingRuntime
metadata:
name: kserve-paddleserver
spec:
containers:
- args:
- --model_name={{.Name}}
- --model_dir=/mnt/models
- --http_port=8080
image: kserve/paddleserver:v0.8.0
name: kserve-container
resources:
limits:
cpu: "1"
memory: 2Gi
requests:
cpu: "1"
memory: 2Gi
supportedModelFormats:
- autoSelect: true
name: paddle
version: "2"
---
apiVersion: serving.kserve.io/v1alpha1
kind: ClusterServingRuntime
metadata:
name: kserve-pmmlserver
spec:
containers:
- args:
- --model_name={{.Name}}
- --model_dir=/mnt/models
- --http_port=8080
image: kserve/pmmlserver:v0.8.0
name: kserve-container
resources:
limits:
cpu: "1"
memory: 2Gi
requests:
cpu: "1"
memory: 2Gi
supportedModelFormats:
- autoSelect: true
name: pmml
version: "3"
- autoSelect: true
name: pmml
version: "4"
---
apiVersion: serving.kserve.io/v1alpha1
kind: ClusterServingRuntime
metadata:
name: kserve-sklearnserver
spec:
containers:
- args:
- --model_name={{.Name}}
- --model_dir=/mnt/models
- --http_port=8080
image: kserve/sklearnserver:v0.8.0
name: kserve-container
resources:
limits:
cpu: "1"
memory: 2Gi
requests:
cpu: "1"
memory: 2Gi
supportedModelFormats:
- autoSelect: true
name: sklearn
version: "1"
---
apiVersion: serving.kserve.io/v1alpha1
kind: ClusterServingRuntime
metadata:
name: kserve-tensorflow-serving
spec:
containers:
- args:
- --model_name={{.Name}}
- --port=9000
- --rest_api_port=8080
- --model_base_path=/mnt/models
- --rest_api_timeout_in_ms=60000
command:
- /usr/bin/tensorflow_model_server
image: tensorflow/serving:2.6.2
name: kserve-container
resources:
limits:
cpu: "1"
memory: 2Gi
requests:
cpu: "1"
memory: 2Gi
supportedModelFormats:
- autoSelect: true
name: tensorflow
version: "1"
- autoSelect: true
name: tensorflow
version: "2"
---
apiVersion: serving.kserve.io/v1alpha1
kind: ClusterServingRuntime
metadata:
name: kserve-torchserve
spec:
containers:
- args:
- torchserve
- --start
- --model-store=/mnt/models/model-store
- --ts-config=/mnt/models/config/config.properties
env:
- name: TS_SERVICE_ENVELOPE
value: '{{.Labels.serviceEnvelope}}'
image: kserve/torchserve-kfs:0.5.3
name: kserve-container
resources:
limits:
cpu: "1"
memory: 2Gi
requests:
cpu: "1"
memory: 2Gi
supportedModelFormats:
- autoSelect: true
name: pytorch
version: "1"
---
apiVersion: serving.kserve.io/v1alpha1
kind: ClusterServingRuntime
metadata:
name: kserve-tritonserver
spec:
containers:
- args:
- tritonserver
- --model-store=/mnt/models
- --grpc-port=9000
- --http-port=8080
- --allow-grpc=true
- --allow-http=true
image: nvcr.io/nvidia/tritonserver:21.09-py3
name: kserve-container
resources:
limits:
cpu: "1"
memory: 2Gi
requests:
cpu: "1"
memory: 2Gi
supportedModelFormats:
- name: tensorrt
version: "8"
- name: tensorflow
version: "1"
- name: tensorflow
version: "2"
- autoSelect: true
name: onnx
version: "1"
- name: pytorch
version: "1"
- autoSelect: true
name: triton
version: "2"
---
apiVersion: serving.kserve.io/v1alpha1
kind: ClusterServingRuntime
metadata:
name: kserve-xgbserver
spec:
containers:
- args:
- --model_name={{.Name}}
- --model_dir=/mnt/models
- --http_port=8080
- --nthread=1
image: kserve/xgbserver:v0.8.0
name: kserve-container
resources:
limits:
cpu: "1"
memory: 2Gi
requests:
cpu: "1"
memory: 2Gi
supportedModelFormats:
- autoSelect: true
name: xgboost
version: "1"
Loading

0 comments on commit a66e7d1

Please sign in to comment.