diff --git a/helm-charts/Makefile b/helm-charts/Makefile index fbd87a9c49..65e612480b 100644 --- a/helm-charts/Makefile +++ b/helm-charts/Makefile @@ -7,13 +7,13 @@ build_all: push_all: gsutil cp build/index.yaml build/*.tgz gs://seldon-charts/ -docs: +docs: # NOTE: helm-docs will ignore the charts listed in the .helmdocsignore file # (in the repository root) helm-docs -o README.md -t ../README.md.tpl install: - GO111MODULE=on go get github.com/norwoodj/helm-docs/cmd/helm-docs@f66fdbd6fe + go install github.com/norwoodj/helm-docs/cmd/helm-docs@f66fdbd6fe package: mkdir -p build && \ @@ -32,4 +32,3 @@ update_index: clean: rm -rfv build - diff --git a/helm-charts/seldon-abtest/README.md b/helm-charts/seldon-abtest/README.md index 8b92ce84c3..920a68dceb 100644 --- a/helm-charts/seldon-abtest/README.md +++ b/helm-charts/seldon-abtest/README.md @@ -1,9 +1,31 @@ # seldon-abtest -![Version: 0.2.0](https://img.shields.io/badge/Version-0.2.0-informational?style=flat-square) +![Version: 0.2.0](https://img.shields.io/static/v1?label=Version&message=0.2.0&color=informational&style=flat-square) Chart to deploy an AB test in Seldon Core v1. Allows you to split traffic between two models. +## Usage + +To use this chart, you will first need to add the `seldonio` Helm repo: + +```bash +helm repo add seldonio https://storage.googleapis.com/seldon-charts +helm repo update +``` + +Once that's done, you should then be able to use the inference graph template as: + +```bash +helm template $MY_MODEL_NAME seldonio/seldon-abtest --namespace $MODELS_NAMESPACE +``` + +Note that you can also deploy the inference graph directly to your cluster +using: + +```bash +helm install $MY_MODEL_NAME seldonio/seldon-abtest --namespace $MODELS_NAMESPACE +``` + **Homepage:** ## Source Code @@ -16,10 +38,10 @@ Chart to deploy an AB test in Seldon Core v1. Allows you to split traffic betwee | Key | Type | Default | Description | |-----|------|---------|-------------| | modela.image.name | string | `"seldonio/mock_classifier"` | | -| modela.image.version | string | `"1.18.0"` | | +| modela.image.version | string | `"1.19.0-dev"` | | | modela.name | string | `"classifier-1"` | | | modelb.image.name | string | `"seldonio/mock_classifier"` | | -| modelb.image.version | string | `"1.18.0"` | | +| modelb.image.version | string | `"1.19.0-dev"` | | | modelb.name | string | `"classifier-2"` | | | predictor.name | string | `"default"` | | | replicas | int | `1` | | diff --git a/helm-charts/seldon-benchmark-workflow/README.md b/helm-charts/seldon-benchmark-workflow/README.md index 346a1c2f67..f1dafc99f5 100644 --- a/helm-charts/seldon-benchmark-workflow/README.md +++ b/helm-charts/seldon-benchmark-workflow/README.md @@ -1,9 +1,31 @@ # seldon-benchmark-workflow -![Version: 0.1](https://img.shields.io/badge/Version-0.1-informational?style=flat-square) +![Version: 0.1](https://img.shields.io/static/v1?label=Version&message=0.1&color=informational&style=flat-square) Seldon Core v1 Benchmark Workflow +## Usage + +To use this chart, you will first need to add the `seldonio` Helm repo: + +```bash +helm repo add seldonio https://storage.googleapis.com/seldon-charts +helm repo update +``` + +Once that's done, you should then be able to use the inference graph template as: + +```bash +helm template $MY_MODEL_NAME seldonio/seldon-benchmark-workflow --namespace $MODELS_NAMESPACE +``` + +Note that you can also deploy the inference graph directly to your cluster +using: + +```bash +helm install $MY_MODEL_NAME seldonio/seldon-benchmark-workflow --namespace $MODELS_NAMESPACE +``` + ## Source Code * diff --git a/helm-charts/seldon-core-analytics/README.md b/helm-charts/seldon-core-analytics/README.md index b4f4df33d6..4696c14396 100644 --- a/helm-charts/seldon-core-analytics/README.md +++ b/helm-charts/seldon-core-analytics/README.md @@ -1,11 +1,27 @@ # seldon-core-analytics -![Version: 1.18.0](https://img.shields.io/badge/Version-1.18.0-informational?style=flat-square) ![AppVersion: 1.18.0](https://img.shields.io/badge/AppVersion-1.18.0-informational?style=flat-square) +![Version: 1.19.0-dev](https://img.shields.io/static/v1?label=Version&message=1.19.0--dev&color=informational&style=flat-square) Prometheus and Grafana installation with a basic Grafana dashboard showing the default Prometheus metrics exposed by Seldon for each inference graph deployed. +## Usage + +To use this chart, you will first need to add the `seldonio` Helm repo: + +```bash +helm repo add seldonio https://storage.googleapis.com/seldon-charts +helm repo update +``` + +Once that's done, you should then be able to deploy the chart as: + +```bash +kubectl create namespace seldon-system +helm install seldon-core-analytics seldonio/seldon-core-analytics --namespace seldon-system +``` + **Homepage:** ## Source Code diff --git a/helm-charts/seldon-core-loadtesting/README.md b/helm-charts/seldon-core-loadtesting/README.md index 0829ec63ab..1fcf10cee6 100644 --- a/helm-charts/seldon-core-loadtesting/README.md +++ b/helm-charts/seldon-core-loadtesting/README.md @@ -1,9 +1,25 @@ # seldon-core-loadtesting -![Version: 0.2.0](https://img.shields.io/badge/Version-0.2.0-informational?style=flat-square) +![Version: 0.2.0](https://img.shields.io/static/v1?label=Version&message=0.2.0&color=informational&style=flat-square) Loadtesting for Seldon Core v1 +## Usage + +To use this chart, you will first need to add the `seldonio` Helm repo: + +```bash +helm repo add seldonio https://storage.googleapis.com/seldon-charts +helm repo update +``` + +Once that's done, you should then be able to deploy the chart as: + +```bash +kubectl create namespace seldon-system +helm install seldon-core-loadtesting seldonio/seldon-core-loadtesting --namespace seldon-system +``` + **Homepage:** ## Source Code diff --git a/helm-charts/seldon-core-operator/README.md b/helm-charts/seldon-core-operator/README.md index 74dee83974..24dab23498 100644 --- a/helm-charts/seldon-core-operator/README.md +++ b/helm-charts/seldon-core-operator/README.md @@ -1,9 +1,25 @@ # seldon-core-operator -![Version: 1.18.0](https://img.shields.io/badge/Version-1.18.0-informational?style=flat-square) ![AppVersion: 1.18.0](https://img.shields.io/badge/AppVersion-1.18.0-informational?style=flat-square) +![Version: 1.19.0-dev](https://img.shields.io/static/v1?label=Version&message=1.19.0--dev&color=informational&style=flat-square) Seldon Core CRD and controller helm chart for Kubernetes. +## Usage + +To use this chart, you will first need to add the `seldonio` Helm repo: + +```bash +helm repo add seldonio https://storage.googleapis.com/seldon-charts +helm repo update +``` + +Once that's done, you should then be able to deploy the chart as: + +```bash +kubectl create namespace seldon-system +helm install seldon-core-operator seldonio/seldon-core-operator --namespace seldon-system +``` + ## Source Code * @@ -31,7 +47,7 @@ Seldon Core CRD and controller helm chart for Kubernetes. | executor.image.pullPolicy | string | `"IfNotPresent"` | | | executor.image.registry | string | `"docker.io"` | | | executor.image.repository | string | `"seldonio/seldon-core-executor"` | | -| executor.image.tag | string | `"1.18.0"` | | +| executor.image.tag | string | `"1.19.0-dev"` | | | executor.metricsPortName | string | `"metrics"` | | | executor.port | int | `8000` | | | executor.prometheus.path | string | `"/prometheus"` | | @@ -44,13 +60,13 @@ Seldon Core CRD and controller helm chart for Kubernetes. | executor.resources.memoryRequest | string | `"512Mi"` | | | executor.serviceAccount.name | string | `"default"` | | | executor.user | int | `8888` | | -| explainer.image | string | `"seldonio/alibiexplainer:1.18.0"` | | +| explainer.image | string | `"seldonio/alibiexplainer:1.19.0-dev"` | | | explainer.image_v2 | string | `"seldonio/mlserver:1.3.5-alibi-explain"` | | | hostNetwork | bool | `false` | | | image.pullPolicy | string | `"IfNotPresent"` | | | image.registry | string | `"docker.io"` | | | image.repository | string | `"seldonio/seldon-core-operator"` | | -| image.tag | string | `"1.18.0"` | | +| image.tag | string | `"1.19.0-dev"` | | | istio.enabled | bool | `false` | | | istio.gateway | string | `"istio-system/seldon-gateway"` | | | istio.tlsMode | string | `""` | | @@ -80,23 +96,23 @@ Seldon Core CRD and controller helm chart for Kubernetes. | predictiveUnit.metricsPortName | string | `"metrics"` | | | predictor_servers.HUGGINGFACE_SERVER.protocols.v2.defaultImageVersion | string | `"1.3.5-huggingface"` | | | predictor_servers.HUGGINGFACE_SERVER.protocols.v2.image | string | `"seldonio/mlserver"` | | -| predictor_servers.MLFLOW_SERVER.protocols.seldon.defaultImageVersion | string | `"1.18.0"` | | +| predictor_servers.MLFLOW_SERVER.protocols.seldon.defaultImageVersion | string | `"1.19.0-dev"` | | | predictor_servers.MLFLOW_SERVER.protocols.seldon.image | string | `"seldonio/mlflowserver"` | | | predictor_servers.MLFLOW_SERVER.protocols.v2.defaultImageVersion | string | `"1.3.5-mlflow"` | | | predictor_servers.MLFLOW_SERVER.protocols.v2.image | string | `"seldonio/mlserver"` | | -| predictor_servers.SKLEARN_SERVER.protocols.seldon.defaultImageVersion | string | `"1.18.0"` | | +| predictor_servers.SKLEARN_SERVER.protocols.seldon.defaultImageVersion | string | `"1.19.0-dev"` | | | predictor_servers.SKLEARN_SERVER.protocols.seldon.image | string | `"seldonio/sklearnserver"` | | | predictor_servers.SKLEARN_SERVER.protocols.v2.defaultImageVersion | string | `"1.3.5-sklearn"` | | | predictor_servers.SKLEARN_SERVER.protocols.v2.image | string | `"seldonio/mlserver"` | | | predictor_servers.TEMPO_SERVER.protocols.v2.defaultImageVersion | string | `"1.3.5-slim"` | | | predictor_servers.TEMPO_SERVER.protocols.v2.image | string | `"seldonio/mlserver"` | | -| predictor_servers.TENSORFLOW_SERVER.protocols.seldon.defaultImageVersion | string | `"1.18.0"` | | +| predictor_servers.TENSORFLOW_SERVER.protocols.seldon.defaultImageVersion | string | `"1.19.0-dev"` | | | predictor_servers.TENSORFLOW_SERVER.protocols.seldon.image | string | `"seldonio/tfserving-proxy"` | | | predictor_servers.TENSORFLOW_SERVER.protocols.tensorflow.defaultImageVersion | string | `"2.1.0"` | | | predictor_servers.TENSORFLOW_SERVER.protocols.tensorflow.image | string | `"tensorflow/serving"` | | | predictor_servers.TRITON_SERVER.protocols.v2.defaultImageVersion | string | `"21.08-py3"` | | | predictor_servers.TRITON_SERVER.protocols.v2.image | string | `"nvcr.io/nvidia/tritonserver"` | | -| predictor_servers.XGBOOST_SERVER.protocols.seldon.defaultImageVersion | string | `"1.18.0"` | | +| predictor_servers.XGBOOST_SERVER.protocols.seldon.defaultImageVersion | string | `"1.19.0-dev"` | | | predictor_servers.XGBOOST_SERVER.protocols.seldon.image | string | `"seldonio/xgboostserver"` | | | predictor_servers.XGBOOST_SERVER.protocols.v2.defaultImageVersion | string | `"1.3.5-xgboost"` | | | predictor_servers.XGBOOST_SERVER.protocols.v2.image | string | `"seldonio/mlserver"` | | @@ -107,7 +123,7 @@ Seldon Core CRD and controller helm chart for Kubernetes. | singleNamespace | bool | `false` | | | storageInitializer.cpuLimit | string | `"1"` | | | storageInitializer.cpuRequest | string | `"100m"` | | -| storageInitializer.image | string | `"seldonio/rclone-storage-initializer:1.18.0"` | | +| storageInitializer.image | string | `"seldonio/rclone-storage-initializer:1.19.0-dev"` | | | storageInitializer.memoryLimit | string | `"1Gi"` | | | storageInitializer.memoryRequest | string | `"100Mi"` | | | usageMetrics.enabled | bool | `false` | | diff --git a/helm-charts/seldon-mab/README.md b/helm-charts/seldon-mab/README.md index 17b7990ef7..781a1b58c1 100644 --- a/helm-charts/seldon-mab/README.md +++ b/helm-charts/seldon-mab/README.md @@ -1,11 +1,33 @@ # seldon-mab -![Version: 0.2.0](https://img.shields.io/badge/Version-0.2.0-informational?style=flat-square) +![Version: 0.2.0](https://img.shields.io/static/v1?label=Version&message=0.2.0&color=informational&style=flat-square) Chart to deploy a multi-armed bandits router over two Seldon Core v1 deployments, so that traffic is sent to the best performing model. You will need to utilize both the `predict` and `send_feedback` API methods. +## Usage + +To use this chart, you will first need to add the `seldonio` Helm repo: + +```bash +helm repo add seldonio https://storage.googleapis.com/seldon-charts +helm repo update +``` + +Once that's done, you should then be able to use the inference graph template as: + +```bash +helm template $MY_MODEL_NAME seldonio/seldon-mab --namespace $MODELS_NAMESPACE +``` + +Note that you can also deploy the inference graph directly to your cluster +using: + +```bash +helm install $MY_MODEL_NAME seldonio/seldon-mab --namespace $MODELS_NAMESPACE +``` + **Homepage:** ## Source Code @@ -25,17 +47,17 @@ You will need to utilize both the `predict` and `send_feedback` API methods. | mab.branches | int | `2` | | | mab.epsilon | float | `0.2` | | | mab.image.name | string | `"seldonio/mab_epsilon_greedy"` | | -| mab.image.version | string | `"1.18.0"` | | +| mab.image.version | string | `"1.19.0-dev"` | | | mab.name | string | `"eg-router"` | | | mab.verbose | int | `1` | | | modela.image.name | string | `"seldonio/mock_classifier"` | | -| modela.image.version | string | `"1.18.0"` | | +| modela.image.version | string | `"1.19.0-dev"` | | | modela.name | string | `"classifier-1"` | | | modelb.image.name | string | `"seldonio/mock_classifier"` | | -| modelb.image.version | string | `"1.18.0"` | | +| modelb.image.version | string | `"1.19.0-dev"` | | | modelb.name | string | `"classifier-2"` | | | predictor.name | string | `"default"` | | | predictorLabels.fluentd | string | `"true"` | | -| predictorLabels.version | string | `"1.18.0"` | | +| predictorLabels.version | string | `"1.19.0-dev"` | | | replicas | int | `1` | | | sdepLabels.app | string | `"seldon"` | | diff --git a/helm-charts/seldon-od-model/README.md b/helm-charts/seldon-od-model/README.md index 7f26b5ee84..e82632feb9 100644 --- a/helm-charts/seldon-od-model/README.md +++ b/helm-charts/seldon-od-model/README.md @@ -1,9 +1,31 @@ # seldon-od-model -![Version: 0.2.0](https://img.shields.io/badge/Version-0.2.0-informational?style=flat-square) +![Version: 0.2.0](https://img.shields.io/static/v1?label=Version&message=0.2.0&color=informational&style=flat-square) Chart to deploy an outlier detector as a single model with Seldon Core v1. +## Usage + +To use this chart, you will first need to add the `seldonio` Helm repo: + +```bash +helm repo add seldonio https://storage.googleapis.com/seldon-charts +helm repo update +``` + +Once that's done, you should then be able to use the inference graph template as: + +```bash +helm template $MY_MODEL_NAME seldonio/seldon-od-model --namespace $MODELS_NAMESPACE +``` + +Note that you can also deploy the inference graph directly to your cluster +using: + +```bash +helm install $MY_MODEL_NAME seldonio/seldon-od-model --namespace $MODELS_NAMESPACE +``` + **Homepage:** ## Source Code diff --git a/helm-charts/seldon-od-transformer/README.md b/helm-charts/seldon-od-transformer/README.md index 00a28fabba..922c3a277e 100644 --- a/helm-charts/seldon-od-transformer/README.md +++ b/helm-charts/seldon-od-transformer/README.md @@ -1,9 +1,31 @@ # seldon-od-transformer -![Version: 0.2.0](https://img.shields.io/badge/Version-0.2.0-informational?style=flat-square) +![Version: 0.2.0](https://img.shields.io/static/v1?label=Version&message=0.2.0&color=informational&style=flat-square) Chart to deploy an outlier detector as a transformer in a Seldon Core v1 inference graph. +## Usage + +To use this chart, you will first need to add the `seldonio` Helm repo: + +```bash +helm repo add seldonio https://storage.googleapis.com/seldon-charts +helm repo update +``` + +Once that's done, you should then be able to use the inference graph template as: + +```bash +helm template $MY_MODEL_NAME seldonio/seldon-od-transformer --namespace $MODELS_NAMESPACE +``` + +Note that you can also deploy the inference graph directly to your cluster +using: + +```bash +helm install $MY_MODEL_NAME seldonio/seldon-od-transformer --namespace $MODELS_NAMESPACE +``` + **Homepage:** ## Source Code diff --git a/helm-charts/seldon-openvino/README.md b/helm-charts/seldon-openvino/README.md index 3fe2ba50f9..d9b08b14ad 100644 --- a/helm-charts/seldon-openvino/README.md +++ b/helm-charts/seldon-openvino/README.md @@ -2,10 +2,32 @@ > **:exclamation: This Helm Chart is deprecated!** -![Version: 0.1](https://img.shields.io/badge/Version-0.1-informational?style=flat-square) +![Version: 0.1](https://img.shields.io/static/v1?label=Version&message=0.1&color=informational&style=flat-square) Proxy integration to deploy models optimized for Intel OpenVINO in Seldon Core v1 +## Usage + +To use this chart, you will first need to add the `seldonio` Helm repo: + +```bash +helm repo add seldonio https://storage.googleapis.com/seldon-charts +helm repo update +``` + +Once that's done, you should then be able to use the inference graph template as: + +```bash +helm template $MY_MODEL_NAME seldonio/seldon-openvino --namespace $MODELS_NAMESPACE +``` + +Note that you can also deploy the inference graph directly to your cluster +using: + +```bash +helm install $MY_MODEL_NAME seldonio/seldon-openvino --namespace $MODELS_NAMESPACE +``` + ## Source Code * diff --git a/helm-charts/seldon-single-model/README.md b/helm-charts/seldon-single-model/README.md index 5a608ba594..bacd187a33 100644 --- a/helm-charts/seldon-single-model/README.md +++ b/helm-charts/seldon-single-model/README.md @@ -1,9 +1,31 @@ # seldon-single-model -![Version: 0.2.0](https://img.shields.io/badge/Version-0.2.0-informational?style=flat-square) +![Version: 0.2.0](https://img.shields.io/static/v1?label=Version&message=0.2.0&color=informational&style=flat-square) Chart to deploy a machine learning model in Seldon Core v1. +## Usage + +To use this chart, you will first need to add the `seldonio` Helm repo: + +```bash +helm repo add seldonio https://storage.googleapis.com/seldon-charts +helm repo update +``` + +Once that's done, you should then be able to use the inference graph template as: + +```bash +helm template $MY_MODEL_NAME seldonio/seldon-single-model --namespace $MODELS_NAMESPACE +``` + +Note that you can also deploy the inference graph directly to your cluster +using: + +```bash +helm install $MY_MODEL_NAME seldonio/seldon-single-model --namespace $MODELS_NAMESPACE +``` + **Homepage:** ## Source Code