A Helm chart for CLIP-as-service, which is a low-latency high-scalability service for embedding images and text. It can be easily integrated as a microservice into neural search solutions.
$ helm repo add jina https://jina.ai/helm-charts/
$ helm install my-clip-release jina/cas-helm
Once this chart is deployed, you can upgrade the deployment via
$ helm upgrade --cleanup-on-fail --install my-clip-release jina/cas-helm
This chart bootstraps a CLIP-as-service deployment on a Kubernetes cluster using the Helm package manager.
To install the chart with the release name my-clip-release on the namespace jina-clip-on-k8s:
$ helm install my-clip-release jina/cas-helm
The command deploys CLIP-as-service on the Kubernetes cluster in the default configuration. The Parameters section lists the parameters that can be configured during installation.
Tip: List all releases using
helm list
To uninstall/delete the my-release deployment:
$ helm delete my-clip-release
| Key | Type | Default | Description |
|---|---|---|---|
| fullnameOverride | string | "" |
|
| image.pullPolicy | string | "IfNotPresent" |
See the kubernetes docs |
| image.repository | string | "jinaai/clip_executor" |
Image to use for deploying, must support ENTRYPOINT[ “jina”, “executor” ] |
| image.tag | string | "master" |
Tag of the image to use |
| imagePullSecrets | list | [] |
|
| gateway.port | int | 51000 |
|
| gateway.protocol | string | grpc |
|
| replicas | int | 1 |
Number of replicas of clip inference executor to serve |
| modelName | string | "Vit-B/32" |
Model name to use for inference execut, default is ViT-B/32. Support all OpenAI released pretrained models |
| deviceName | string | "cpu" |
cuda or cpu. Default is None means auto-detect. |
| nameOverride | string | "" |
|
| nodeSelector | object | {} |
See the Kubernetes docs |
| podAnnotations | object | {} |
Extra annotations for all pods |
| podSecurityContext | object | {} |
Security context for the pods. The default container can run as any user/group and does not run with elevated |
| prometheus.expose | bool | false |
If true, prometheus metrics are exposed on /metrics |
| resources | object | {} |
Resource limits and requests for the mlflow pods |
| securityContext | object | {} |
Security context for the containers. Take presedence over podSecurityContext. |
| service.port | int | 5000 |
|
| service.type | string | "ClusterIP" |
See the Kubernetes docs |
| serviceAccount.annotations | object | {} |
Annotations to add to the service account |
| serviceAccount.create | bool | true |
Specifies whether a service account should be created |
| serviceAccount.name | string | "" |
The name of the service account to use. If not set and create is true, a name is generated using the fullname template |
| tolerations | list | [] |
See the Kubernetes docs |
Specify each parameter using the --set key=value[,key=value] argument to helm install. For example,
$ helm install my-clip-release \
--set modelName=ViT-B/16 \
jina/cas-helm
The above command sets the pretrained CLIP model name modelName to ViT-B/16.
Alternatively, a YAML file that specifies the values for the parameters can be provided while installing the chart. For example,
helm install my-clip-release -f values.yaml jina/cas-helm
Tip: You can use the default values.yaml
Copyright 2020-2022 Jina AI Limited. All rights reserved.
Licensed under the Apache License, Version 2.0 (the “License”); you may not use this file except in compliance with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an “AS IS” BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.