v1alpha1
API reference
ApplicationLogsParams describes a set of parameters that can be used to build a search query for an application's logs. In most cases, only ContainerId, From, and Size are required in order to create a search query.
Application Logs
ErrorDefault Error
ErrorDefault Error
Read the cluster info
Cluster Info
ErrorDefault Error
Read the healthcheck
Version
Version
Read the git diff for an experiment
Name identifies a resource
Namespace provides a logical grouping of resources
Hash
Git Diff
ErrorDefault Error
Read the git commits for an experiment
Name identifies a resource
Namespace provides a logical grouping of resources
Git Logs
ErrorDefault Error
The namespace that this seldon deployment belongs to.
The name of the seldon deployment to get inference logs for.
The predictor name of the seldon deployment to get inference logs for. e.g. "default", "canary", "shadow"
The container name of the seldon deployment to get inference logs for. e.g. "my-container", "input-transformer", "output-transformer"
Inference logs metadata for seldon deployments
Bad Request
Internal Server Error
Service not implemented or enabled
An unexpected error response.
Get inference logs for seldon deployments.
The namespace that this seldon deployment belongs to.
The name of the seldon deployment to get inference logs for.
The predictor name of the seldon deployment to get inference logs for. e.g. "default", "canary", "shadow"
defaultThe container name of the seldon deployment to get inference logs for. e.g. "my-container", "input-transformer", "output-transformer"
my-containerThe start time of the inference logs to get.
2020-01-01T00:00:00ZThe end time of the inference logs to get.
2020-01-01T00:00:00ZThe maximum number of inference logs to get.
100The page token to use to get the next page of inference logs.
my-page-tokenInference logs metadata for seldon deployments
Bad Request
Internal Server Error
Service not implemented or enabled
An unexpected error response.
Get inference logs for seldon models.
The namespace that this seldon deployment belongs to.
The name of the seldon pipeline to get inference logs for.
The name of the seldon model to get inference logs for.
The start time of the inference logs to get.
2020-01-01T00:00:00ZThe end time of the inference logs to get.
2020-01-01T00:00:00ZThe maximum number of inference logs to get.
100The page token to use to get the next page of inference logs.
my-page-tokenInference logs metadata for seldon models
Bad Request
Internal Server Error
Service not implemented or enabled
An unexpected error response.
The namespace that this seldon deployment belongs to.
The name of the seldon pipeline to get inference logs for.
The name of the seldon model to get inference logs for.
Inference logs metadata for seldon models
Bad Request
Internal Server Error
Service not implemented or enabled
An unexpected error response.
List the current health of a specific Seldon Deploy dependency or all of them
OK
Internal Server Error
An unexpected error response.
The namespace to delete secret from.
The secret type of the secret to be deleted, can be one of (bucket, registry).
The name of the secret to delete.
OK
Bad Request
Forbidden
Not Found
Internal Server Error
An unexpected error response.
The namespace to list secrets in.
The secret type, can be one of (bucket, registry) or all to list all secrets used by Seldon Deploy.
Optional filter for the format of bucket secrets. Only applicable when secret_type is bucket.
Optional The maximum number of Secrets to return in the response.
Optional pagination token returned from a previous call to List that indicates where this listing should continue from.
Optional search string for secrets.
OK
Bad Request
Forbidden
Internal Server Error
An unexpected error response.
Creates a registry secret according to specified parameters.
The namespace to create secret in.
The secret name must contain only lowercase letters, digits or dashes, must start and end with a lowercase letter or digit and must be no more than 30 characters in length.
The raw json docker credentials config.
OK
Bad Request
Forbidden
Internal Server Error
An unexpected error response.
Creates a S3 bucket secret according to specified parameters.
The namespace to create secret in.
The remote name must contain only lowercase letters, digits or dashes, must start and end with a lowercase letter or digit and must be no more than 30 characters in length. The created secret will be named {remote}-bucket-envvars or {remote}-bucket-params, for Core 1 or Core 2, respectively.
The version of the secret to use (CORE_1 or CORE_2). Default value is CORE_1
The S3 account credentials to populate the secret. See documentation for how to generate these.
The S3 access key ID.
The S3 secret access key.
OK
Bad Request
Forbidden
Internal Server Error
An unexpected error response.
Creates a generic rclone bucket secret according to specified parameters.
The namespace to create secret in.
The remote name must contain only lowercase letters, digits or dashes, must start and end with a lowercase letter or digit and must be no more than 30 characters in length. The created secret will be named {remote}-bucket-envvars or {remote}-bucket-params, for Core 1 or Core 2, respectively.
The format for the secret. Must be one of (env_var, config_param). env_var formats the secret as environment variables, each with a key and a value. env_var is suitable for Seldon Core v1 and Argo Workflows. config_param formats the secret to be compatible with Rclone's HTTP API. config_param is suitable for Seldon Core v2.
The rclone key value pairs.
The Rclone type of the remote; see the Rclone documentation for available types.
s3OK
Bad Request
Forbidden
Internal Server Error
An unexpected error response.
Creates a GCS bucket secret according to specified parameters.
The namespace to create secret in.
The remote name must contain only lowercase letters, digits or dashes, must start and end with a lowercase letter or digit and must be no more than 30 characters in length. The created secret will be named {remote}-bucket-envvars or {remote}-bucket-params, for Core 1 or Core 2, respectively.
The version of the secret to use (CORE_V1 or CORE_V2). Default value is CORE_V1
The GCS account credentials to populate the secret. See documentation for how to generate this.
OK
Bad Request
Forbidden
Internal Server Error
An unexpected error response.
Endpoint is available only when user management is enabled configured - refer to the docs for how to do this. Sends an email to the user with a link to reset their password. The caller must have `write` permission on `system/iam`.
The username of the user to be deleted.
OK
Bad Request
Forbidden
An unexpected error response.
Endpoint is available only when user management is enabled configured - refer to the docs for how to do this. Add user to a group. The caller must have `write` permission on `system/iam`.
The name of the user to be added to a group.
The name of the group to which to add a user.
OK
Bad Request
Forbidden
An unexpected error response.
Endpoint is available only when user management is enabled configured - refer to the docs for how to do this. Delete user from a group. The caller must have `write` permission on `system/iam`.
The name of the user to be removed from a group.
The name of the group from which to remove a user.
OK
Bad Request
Forbidden
An unexpected error response.
Endpoint is available only when user management is enabled configured - refer to the docs for how to do this. List all groups of a user. The caller must have `read` permission on `system/iam`.
The name of the user for who to get groups.
OK
Bad Request
Forbidden
An unexpected error response.
Endpoint is available only when user management is enabled configured - refer to the docs for how to do this. Delete a user. The caller must have `write` permission on `system/iam`.
The username of the user to be deleted.
OK
Bad Request
Forbidden
An unexpected error response.
Endpoint is available only when user management is enabled configured - refer to the docs for how to do this. Create a user. The caller must have `write` permission on `system/iam`.
The username of the user. It must consist of only letters, numbers, underscores, and dashes, and be at most 30 characters. It can be used for login and be part of policies.
The email of the user. It must be a valid email. The user will receive instructions for setting credentials on this email.
OK
Bad Request
Forbidden
An unexpected error response.
Endpoint is available only when user management is enabled configured - refer to the docs for how to do this. List users. The caller must have `read` permission on `system/iam`.
The username of the user to fetch. It must consist of only letters, numbers, underscores, and dashes, and be at most 30 characters.
The email of the user to fetch. It must be a valid email.
OK
Bad Request
Forbidden
An unexpected error response.
List all users and groups who have access to the given resource/action pair. The user calling this endpoint must have `grant` access to the given resource.
The action a target should be able to perform on the given resource.
The resource a target should have access to.
OK
Bad Request
Forbidden
An unexpected error response.
List all permissions associated with the given users and groups. A regular user will be able to see only their permissions and the permissions of their groups. A user with `read` permission on `system/iam` can see all permissions.
The user IDs for which to list permission.
The groups for which to list permission.
If true will list the permissions of the user making the request. All users have permissions to check their permissions.
OK
Bad Request
Forbidden
An unexpected error response.
Create an authorization policy. The user must have `grant` permissions on the resource in the policy.
OK
Bad Request
Forbidden
An unexpected error response.
Delete an authorization policy. The user must have `grant` permissions on the resource in the policy.
The user IDs from which to revoke the given resource/action permission.
The groups from which to revoke the given resource/action permission.
The action part of the resource/action permission to revoke.
The resource part of the resource/action permission to revoke.
OK
Bad Request
Forbidden
An unexpected error response.
Endpoint is available only when user management is enabled configured - refer to the docs for how to do this. Delete a group. The caller must have `write` permission on `system/iam`.
The name of the group to be deleted.
OK
Bad Request
Forbidden
An unexpected error response.
Endpoint is available only when user management is enabled configured - refer to the docs for how to do this. List all members of a group. The caller must have `read` permission on `system/iam`.
The name of the group for which to get members.
OK
Bad Request
Forbidden
An unexpected error response.
Endpoint is available only when user management is enabled configured - refer to the docs for how to do this. Create a group. The caller must have `write` permission on `system/iam`.
The name of the group to be created.
OK
Bad Request
Forbidden
An unexpected error response.
Endpoint is available only when user management is enabled configured - refer to the docs for how to do this. List all groups. The caller must have `read` permission on `system/iam`.
OK
Bad Request
Forbidden
An unexpected error response.
Optional. The maximum number of Folders to return in the response.
Optional. A pagination token returned from a previous call to List
that indicates where this listing should continue from.
Optional. Can be used to specify which fields of RuntimeMetadata you wish to return in the response. If left empty all fields will be returned.
OK
No Content
Bad Request
Internal Server Error
An unexpected error response.
The URI for the storage bucket containing the model, or the URI to the docker image for custom models. It must be a valid URI as defined in RFC 3986, and must not exceed 200 characters.
gs://seldon-models/sklearn/irisThe name of the model. It must not exceed 200 characters.
Iris ClassifierThe version of the model. It must not exceed 50 characters.
"v0.0.1"Example: v1.2.3The artifact type of the model. This is the library used to develop the model.
UNKNOWNExample: SKLEARNPossible values: The task type of the model. It must not exceed 50 characters.
classificationThe creation timestamp for the model metadata entry. It is automatically created by the Metadata service and cannot be modified. The timestamp is using the RFC 3339 format/
2017-01-15T01:30:15.01ZThe project that this model belongs to.
project_1OK
Bad Request
Internal Server Error
An unexpected error response.
The URI for the storage bucket containing the model, or the URI to the docker image for custom models. It must be a valid URI as defined in RFC 3986, and must not exceed 200 characters.
gs://seldon-models/sklearn/irisThe name of the model. It must not exceed 200 characters.
Iris ClassifierThe version of the model. It must not exceed 50 characters.
"v0.0.1"Example: v1.2.3The artifact type of the model. This is the library used to develop the model.
UNKNOWNExample: SKLEARNPossible values: The task type of the model. It must not exceed 50 characters.
classificationThe creation timestamp for the model metadata entry. It is automatically created by the Metadata service and cannot be modified. The timestamp is using the RFC 3339 format/
2017-01-15T01:30:15.01ZThe project that this model belongs to.
project_1OK
Bad Request
Internal Server Error
An unexpected error response.
The URI for the storage bucket containing the model, or the URI to the docker image for custom models. It must be a valid URI as defined in RFC 3986, and must not exceed 200 characters.
The project that this model belongs to.
OK
No Content
Internal Server Error
An unexpected error response.
List takes several parameters that are present in the Model Metadata and tries to list all metadata entries that match all supplied fields. To filter by tags or metrics you can use a map as a query parameter. For example: ?tags[key]=value.
For more complex queries where other logical operators like OR, NOT, etc.
Optional. The maximum number of Folders to return in the response.
Optional. A pagination token returned from a previous call to List
that indicates where this listing should continue from.
Optional. Can be used to specify which fields of Model you wish to return in the response. If left empty all fields will be returned.
Based on https://cloud.google.com/apis/design/design_patterns#sorting_order
The order in which to return the model metadata. The string value should follow SQL syntax: comma separated list of fields. The default sorting order is ascending. To specify descending order for a field, a suffix " desc" should be appended to the field name. Valid field names include: uri, name, version, project, artifact_type, task_type.
OK
No Content
Bad Request
Internal Server Error
An unexpected error response.
Read the version
Version
Version
Read the request user
User
User
Validate the given Seldon Deployment
Namespace provides a logical grouping of resources
+genclient +genclient:noStatus +kubebuilder:object:root=true +k8s:deepcopy-gen:interfaces=k8s.io/apimachinery/pkg/runtime.Object +kubebuilder:storageversion SeldonDeployment is the Schema for the seldondeployments API +k8s:openapi-gen=true +kubebuilder:resource:shortName=sdep +kubebuilder:subresource:status +kubebuilder:subresource:scale:specpath=.spec.replicas,statuspath=.status.replicas
APIVersion defines the versioned schema of this representation of an object. Servers should convert recognized schemas to the latest internal value, and may reject unrecognized values. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#resources +optional
Kind is a string value representing the REST resource this object represents. Servers may infer this from the endpoint the client submits requests to. Cannot be updated. In CamelCase. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#types-kinds +optional
GenericMessage Message
ErrorDefault Error
list objects of kind resource for Seldon Deployment
Name identifies a resource
Namespace provides a logical grouping of resources
Component differentiates between types of model (e.g. predictor, explainer... etc)
Resource list
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
Create Seldon Deployment prediction
Name identifies a resource
Namespace provides a logical grouping of resources
PredictionFile
Generic
ErrorDefault Error
No content
Read the specified Seldon Deployment predict curl
Name identifies a resource
Namespace provides a logical grouping of resources
Generic
ErrorDefault Error
No content
Create Seldon Deployment prediction
Name identifies a resource
Namespace provides a logical grouping of resources
Generic
ErrorDefault Error
No content
Read the specified Seldon Deployment Outlier Detector
Name identifies a resource
Namespace provides a logical grouping of resources
Detector Name
Detector configuration data
ErrorDefault Error
Read the specified Seldon Deployment Outlier Detector
Name identifies a resource
Namespace provides a logical grouping of resources
Detector Name
GenericMessage Message
ErrorDefault Error
Create the specified Seldon Deployment Outlier Detector
Name identifies a resource
Namespace provides a logical grouping of resources
Detector configuration data
ErrorDefault Error
ErrorDefault Error
Read the specified Seldon Deployment Outlier Detector
Name identifies a resource
Namespace provides a logical grouping of resources
Detector configuration data list
ErrorDefault Error
Read the specified Seldon Deployment Metrics Server
Name identifies a resource
Namespace provides a logical grouping of resources
Detector Name
Detector configuration data
ErrorDefault Error
Read the specified Seldon Deployment Metrics Server
Name identifies a resource
Namespace provides a logical grouping of resources
Detector Name
GenericMessage Message
ErrorDefault Error
Create the specified Seldon Deployment Metrics Server
Name identifies a resource
Namespace provides a logical grouping of resources
Detector configuration data
ErrorDefault Error
ErrorDefault Error
Read the specified Seldon Deployment Metrics Server
Name identifies a resource
Namespace provides a logical grouping of resources
Detector configuration data list
ErrorDefault Error
Get the specified Seldon Deployment predictions feature statistics
Name identifies a resource
Namespace provides a logical grouping of resources
DeploymentFeatureData represents a deployment level feature data
AggregateOverTime is a boolean to decide if the distribution response is to be aggregated over the time period selected
DeploymentEndpoint refers to endpoint/predictor name relevant to the feature distribution query
DeploymentType refers to kubernetes kind of the deployment relevant to the feature distribution query
DeploymentName refers to name of the deployment relevant to the feature distribution query
DeploymentNamespace refers to namespace of the deployment relevant to the feature distribution query
DeploymentNode refers to node name relevant to the feature distribution query
FeatureName refers to the name of feature as per the model predictions schema
Feature Prediction Interaction
ReferenceData is a boolean to compute stats and distributions over reference data or inference data (false by default)
Feature Statistics Response
ErrorDefault Error
ErrorDefault Error
Get the specified Seldon Deployment predictions feature distribution
Name identifies a resource
Namespace provides a logical grouping of resources
DeploymentFeatureData represents a deployment level feature data
AggregateOverTime is a boolean to decide if the distribution response is to be aggregated over the time period selected
DeploymentEndpoint refers to endpoint/predictor name relevant to the feature distribution query
DeploymentType refers to kubernetes kind of the deployment relevant to the feature distribution query
DeploymentName refers to name of the deployment relevant to the feature distribution query
DeploymentNamespace refers to namespace of the deployment relevant to the feature distribution query
DeploymentNode refers to node name relevant to the feature distribution query
FeatureName refers to the name of feature as per the model predictions schema
Feature Prediction Interaction
ReferenceData is a boolean to compute stats and distributions over reference data or inference data (false by default)
Feature Distribution Response
ErrorDefault Error
ErrorDefault Error
Read the specified Seldon Deployment Drift Detector
Name identifies a resource
Namespace provides a logical grouping of resources
Detector Name
Detector configuration data
ErrorDefault Error
Read the specified Seldon Deployment Drift Detector
Name identifies a resource
Namespace provides a logical grouping of resources
Detector Name
GenericMessage Message
ErrorDefault Error
Create the specified Seldon Deployment Drift Detector
Name identifies a resource
Namespace provides a logical grouping of resources
Detector configuration data
ErrorDefault Error
ErrorDefault Error
Read the specified Seldon Deployment Drift Detector
Name identifies a resource
Namespace provides a logical grouping of resources
Detector configuration data list
ErrorDefault Error
Delete Seldon Deployment load test jobs
JobName identifies a job name
Name identifies a resource
Namespace provides a logical grouping of resources
GenericMessage Message
ErrorDefault Error
Create Seldon Deployment load test
Name identifies a resource
Namespace provides a logical grouping of resources
Prediction contains predict payload
Connections Number provides number of allowed connections
Requests Number provides number of allowed requests
Duration of load test in seconds
GenericMessage Message
ErrorDefault Error
List Seldon Deployment load test jobs
Name identifies a resource
Namespace provides a logical grouping of resources
GenericMessage Message
ErrorDefault Error
Revert the git commit
Name identifies a resource
Namespace provides a logical grouping of resources
Hash
Action
Message
GenericMessage Message
ErrorDefault Error
Restore the git commit
Name identifies a resource
Namespace provides a logical grouping of resources
Hash
Action
Message
GenericMessage Message
ErrorDefault Error
Read the git commits for a Seldon deployment
Name identifies a resource
Namespace provides a logical grouping of resources
Git Logs
ErrorDefault Error
Read the git diff for a Seldon deployment
Name identifies a resource
Namespace provides a logical grouping of resources
Hash
Git Diff
ErrorDefault Error
Create Seldon Deployment explanation
Namespace provides a logical grouping of resources
Name identifies a resource
Generic
ErrorDefault Error
No content
Get details on the seldondeployment's batch job
Name identifies a resource
Namespace provides a logical grouping of resources
JobName identifies a job name
Batch Job Response
ErrorDefault Error
Create the seldondeployment's batch jobs
Name identifies a resource
Namespace provides a logical grouping of resources
BatchJobDefinition is a common interface for BatchJob v1 and v2 which specifies necessary arguments
Batch Data Type (data, json or str)
Interval between batches
Batch Method (predict, feedback (for SCv1))
Batch Payload Type (ndarray, tensor, tftensor, v2raw, v2binary - only if DataType=data)
Number of retries for each instance
Size of the batch (number of predictions per request)
Batch Transport Protocol (rest or grpc)
Number of batch workers
S3 URI of input data file
Container Resources for running batch jobs: limits: CPU
Container Resources for running batch jobs: limits: Memory
name of Kubernetes Secret with S3 credentials
S3 URI of output data file
Size of PVC required for the batch job
Container Resources for running batch jobs: requests: CPU
Container Resources for running batch jobs: requests: Memory
Batch Job Post Response
ErrorDefault Error
ErrorDefault Error
Read the seldondeployment's batch jobs
Name identifies a resource
Namespace provides a logical grouping of resources
Limit of items returned in one response
Requested page
Batch Jobs List Response
ErrorDefault Error
ErrorDefault Error
Update the specified Seldon Deployment
Name identifies a resource
Namespace provides a logical grouping of resources
Action
Message
+genclient +genclient:noStatus +kubebuilder:object:root=true +k8s:deepcopy-gen:interfaces=k8s.io/apimachinery/pkg/runtime.Object +kubebuilder:storageversion SeldonDeployment is the Schema for the seldondeployments API +k8s:openapi-gen=true +kubebuilder:resource:shortName=sdep +kubebuilder:subresource:status +kubebuilder:subresource:scale:specpath=.spec.replicas,statuspath=.status.replicas
APIVersion defines the versioned schema of this representation of an object. Servers should convert recognized schemas to the latest internal value, and may reject unrecognized values. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#resources +optional
Kind is a string value representing the REST resource this object represents. Servers may infer this from the endpoint the client submits requests to. Cannot be updated. In CamelCase. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#types-kinds +optional
Seldon Deployment
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
Read the specified Seldon Deployment
Name identifies a resource
Namespace provides a logical grouping of resources
Seldon Deployment
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
Delete the specified Seldon Deployment
Name identifies a resource
Namespace provides a logical grouping of resources
Action
Message
Generic
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
No content
Create a Seldon Deployment
Namespace provides a logical grouping of resources
Action
Message
+genclient +genclient:noStatus +kubebuilder:object:root=true +k8s:deepcopy-gen:interfaces=k8s.io/apimachinery/pkg/runtime.Object +kubebuilder:storageversion SeldonDeployment is the Schema for the seldondeployments API +k8s:openapi-gen=true +kubebuilder:resource:shortName=sdep +kubebuilder:subresource:status +kubebuilder:subresource:scale:specpath=.spec.replicas,statuspath=.status.replicas
APIVersion defines the versioned schema of this representation of an object. Servers should convert recognized schemas to the latest internal value, and may reject unrecognized values. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#resources +optional
Kind is a string value representing the REST resource this object represents. Servers may infer this from the endpoint the client submits requests to. Cannot be updated. In CamelCase. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#types-kinds +optional
Seldon Deployment
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
list objects of kind Seldon Deployment
Namespace provides a logical grouping of resources
Seldon Deployment list
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
Create Seldon Pipeline prediction
Name identifies a resource
Namespace provides a logical grouping of resources
PredictionFile
Generic
ErrorDefault Error
No content
Read the specified Seldon Pipeline predict curl
Name identifies a resource
Namespace provides a logical grouping of resources
Generic
ErrorDefault Error
No content
Read the specified Seldon Pipeline Outlier Detector
Name identifies a resource
Namespace provides a logical grouping of resources
Detector Name
Detector configuration data
ErrorDefault Error
Read the specified Seldon Pipeline Outlier Detector
Name identifies a resource
Namespace provides a logical grouping of resources
Detector Name
GenericMessage Message
ErrorDefault Error
Create the specified Seldon Pipeline Outlier Detector
Name identifies a resource
Namespace provides a logical grouping of resources
Detector configuration data
ErrorDefault Error
List the specified Seldon Pipeline Outlier Detector
Name identifies a resource
Namespace provides a logical grouping of resources
Detector configuration data list
ErrorDefault Error
Get the specified Seldon Deployment predictions feature statistics
Name identifies a resource
Namespace provides a logical grouping of resources
DeploymentFeatureData represents a deployment level feature data
AggregateOverTime is a boolean to decide if the distribution response is to be aggregated over the time period selected
DeploymentEndpoint refers to endpoint/predictor name relevant to the feature distribution query
DeploymentType refers to kubernetes kind of the deployment relevant to the feature distribution query
DeploymentName refers to name of the deployment relevant to the feature distribution query
DeploymentNamespace refers to namespace of the deployment relevant to the feature distribution query
DeploymentNode refers to node name relevant to the feature distribution query
FeatureName refers to the name of feature as per the model predictions schema
Feature Prediction Interaction
ReferenceData is a boolean to compute stats and distributions over reference data or inference data (false by default)
Feature Statistics Response
ErrorDefault Error
ErrorDefault Error
Get the specified Seldon Deployment predictions feature distribution
Name identifies a resource
Namespace provides a logical grouping of resources
DeploymentFeatureData represents a deployment level feature data
AggregateOverTime is a boolean to decide if the distribution response is to be aggregated over the time period selected
DeploymentEndpoint refers to endpoint/predictor name relevant to the feature distribution query
DeploymentType refers to kubernetes kind of the deployment relevant to the feature distribution query
DeploymentName refers to name of the deployment relevant to the feature distribution query
DeploymentNamespace refers to namespace of the deployment relevant to the feature distribution query
DeploymentNode refers to node name relevant to the feature distribution query
FeatureName refers to the name of feature as per the model predictions schema
Feature Prediction Interaction
ReferenceData is a boolean to compute stats and distributions over reference data or inference data (false by default)
Feature Distribution Response
ErrorDefault Error
ErrorDefault Error
Read the specified Seldon Pipeline Drift Detector
Name identifies a resource
Namespace provides a logical grouping of resources
Detector Name
Detector configuration data
ErrorDefault Error
Read the specified Seldon Pipeline Drift Detector
Name identifies a resource
Namespace provides a logical grouping of resources
Detector Name
GenericMessage Message
ErrorDefault Error
Create the specified Seldon Pipeline Drift Detector
Name identifies a resource
Namespace provides a logical grouping of resources
Detector configuration data
ErrorDefault Error
List the specified Seldon Pipeline Drift Detector
Name identifies a resource
Namespace provides a logical grouping of resources
Detector configuration data list
ErrorDefault Error
List objects of kind Seldon Models within a Pipeline
Name identifies a resource
Namespace provides a logical grouping of resources
Seldon Model List
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
Delete Pipeline load test jobs
JobName identifies a job name
Name identifies a resource
Namespace provides a logical grouping of resources
GenericMessage Message
ErrorDefault Error
Create Pipeline load test
Name identifies a resource
Namespace provides a logical grouping of resources
Prediction contains predict payload
Connections Number provides number of allowed connections
Requests Number provides number of allowed requests
Duration of load test in seconds
GenericMessage Message
ErrorDefault Error
List Pipeline load test jobs
Name identifies a resource
Namespace provides a logical grouping of resources
GenericMessage Message
ErrorDefault Error
Read the git commits for a pipeline
Name identifies a resource
Namespace provides a logical grouping of resources
Git Logs
ErrorDefault Error
Read the git diff for a pipeline
Name identifies a resource
Namespace provides a logical grouping of resources
Hash
Git Diff
ErrorDefault Error
Update the specified Seldon Pipeline Explainer
Name identifies a resource
Namespace provides a logical grouping of resources
Action
Message
Model is the Schema for the models API
APIVersion defines the versioned schema of this representation of an object. Servers should convert recognized schemas to the latest internal value, and may reject unrecognized values. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#resources +optional
CreationTimestamp is a timestamp representing the server time when this object was created. It is not guaranteed to be set in happens-before order across separate operations. Clients may not set this value. It is represented in RFC3339 form and is in UTC.
Populated by the system. Read-only. Null for lists. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#metadata +optional
Number of seconds allowed for this object to gracefully terminate before it will be removed from the system. Only set when deletionTimestamp is also set. May only be shortened. Read-only. +optional
DeletionTimestamp is RFC 3339 date and time at which this resource will be deleted. This field is set by the server when a graceful deletion is requested by the user, and is not directly settable by a client. The resource is expected to be deleted (no longer visible from resource lists, and not reachable by name) after the time in this field, once the finalizers list is empty. As long as the finalizers list contains items, deletion is blocked. Once the deletionTimestamp is set, this value may not be unset or be set further into the future, although it may be shortened or the resource may be deleted prior to this time. For example, a user may request that a pod is deleted in 30 seconds. The Kubelet will react by sending a graceful termination signal to the containers in the pod. After that 30 seconds, the Kubelet will send a hard termination signal (SIGKILL) to the container and after cleanup, remove the pod from the API. In the presence of network partitions, this object may still exist after this timestamp, until an administrator or automated process can determine the resource is fully terminated. If not set, graceful deletion of the object has not been requested.
Populated by the system when a graceful deletion is requested. Read-only. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#metadata +optional
Must be empty before the object is deleted from the registry. Each entry is an identifier for the responsible component that will remove the entry from the list. If the deletionTimestamp of the object is non-nil, entries in this list can only be removed. Finalizers may be processed and removed in any order. Order is NOT enforced because it introduces significant risk of stuck finalizers. finalizers is a shared field, any actor with permission can reorder it. If the finalizer list is processed in order, then this can lead to a situation in which the component responsible for the first finalizer in the list is waiting for a signal (field value, external system, or other) produced by a component responsible for a finalizer later in the list, resulting in a deadlock. Without enforced ordering finalizers are free to order amongst themselves and are not vulnerable to ordering changes in the list. +optional +patchStrategy=merge
GenerateName is an optional prefix, used by the server, to generate a unique name ONLY IF the Name field has not been provided. If this field is used, the name returned to the client will be different than the name passed. This value will also be combined with a unique suffix. The provided value has the same validation rules as the Name field, and may be truncated by the length of the suffix required to make the value unique on the server.
If this field is specified and the generated name exists, the server will return a 409.
Applied only if Name is not specified. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#idempotency +optional
A sequence number representing a specific generation of the desired state. Populated by the system. Read-only. +optional
Kind is a string value representing the REST resource this object represents. Servers may infer this from the endpoint the client submits requests to. Cannot be updated. In CamelCase. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#types-kinds +optional
Name must be unique within a namespace. Is required when creating resources, although some resources may allow a client to request the generation of an appropriate name automatically. Name is primarily intended for creation idempotence and configuration definition. Cannot be updated. More info: https://kubernetes.io/docs/concepts/overview/working-with-objects/names#names +optional
Namespace defines the space within which each name must be unique. An empty namespace is equivalent to the "default" namespace, but "default" is the canonical representation. Not all objects are required to be scoped to a namespace - the value of this field for those objects will be empty.
Must be a DNS_LABEL. Cannot be updated. More info: https://kubernetes.io/docs/concepts/overview/working-with-objects/namespaces +optional
An opaque value that represents the internal version of this object that can be used by clients to determine when objects have changed. May be used for optimistic concurrency, change detection, and the watch operation on a resource or set of resources. Clients must treat these values as opaque and passed unmodified back to the server. They may only be valid for a particular resource or set of resources.
Populated by the system. Read-only. Value must be treated as opaque by clients and . More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#concurrency-control-and-consistency +optional
Deprecated: selfLink is a legacy read-only field that is no longer populated by the system. +optional
UID is a type that holds unique ID values, including UUIDs. Because we don't ONLY use UUIDs, this is an alias to string. Being a type captures intent and helps make sure that UIDs and names do not get conflated.
Seldon Model
ErrorDefault Error
Read the specified Seldon Pipeline Explainer
Name identifies a resource
Namespace provides a logical grouping of resources
Action
Message
Model is the Schema for the models API
APIVersion defines the versioned schema of this representation of an object. Servers should convert recognized schemas to the latest internal value, and may reject unrecognized values. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#resources +optional
CreationTimestamp is a timestamp representing the server time when this object was created. It is not guaranteed to be set in happens-before order across separate operations. Clients may not set this value. It is represented in RFC3339 form and is in UTC.
Populated by the system. Read-only. Null for lists. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#metadata +optional
Number of seconds allowed for this object to gracefully terminate before it will be removed from the system. Only set when deletionTimestamp is also set. May only be shortened. Read-only. +optional
DeletionTimestamp is RFC 3339 date and time at which this resource will be deleted. This field is set by the server when a graceful deletion is requested by the user, and is not directly settable by a client. The resource is expected to be deleted (no longer visible from resource lists, and not reachable by name) after the time in this field, once the finalizers list is empty. As long as the finalizers list contains items, deletion is blocked. Once the deletionTimestamp is set, this value may not be unset or be set further into the future, although it may be shortened or the resource may be deleted prior to this time. For example, a user may request that a pod is deleted in 30 seconds. The Kubelet will react by sending a graceful termination signal to the containers in the pod. After that 30 seconds, the Kubelet will send a hard termination signal (SIGKILL) to the container and after cleanup, remove the pod from the API. In the presence of network partitions, this object may still exist after this timestamp, until an administrator or automated process can determine the resource is fully terminated. If not set, graceful deletion of the object has not been requested.
Populated by the system when a graceful deletion is requested. Read-only. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#metadata +optional
Must be empty before the object is deleted from the registry. Each entry is an identifier for the responsible component that will remove the entry from the list. If the deletionTimestamp of the object is non-nil, entries in this list can only be removed. Finalizers may be processed and removed in any order. Order is NOT enforced because it introduces significant risk of stuck finalizers. finalizers is a shared field, any actor with permission can reorder it. If the finalizer list is processed in order, then this can lead to a situation in which the component responsible for the first finalizer in the list is waiting for a signal (field value, external system, or other) produced by a component responsible for a finalizer later in the list, resulting in a deadlock. Without enforced ordering finalizers are free to order amongst themselves and are not vulnerable to ordering changes in the list. +optional +patchStrategy=merge
GenerateName is an optional prefix, used by the server, to generate a unique name ONLY IF the Name field has not been provided. If this field is used, the name returned to the client will be different than the name passed. This value will also be combined with a unique suffix. The provided value has the same validation rules as the Name field, and may be truncated by the length of the suffix required to make the value unique on the server.
If this field is specified and the generated name exists, the server will return a 409.
Applied only if Name is not specified. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#idempotency +optional
A sequence number representing a specific generation of the desired state. Populated by the system. Read-only. +optional
Kind is a string value representing the REST resource this object represents. Servers may infer this from the endpoint the client submits requests to. Cannot be updated. In CamelCase. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#types-kinds +optional
Name must be unique within a namespace. Is required when creating resources, although some resources may allow a client to request the generation of an appropriate name automatically. Name is primarily intended for creation idempotence and configuration definition. Cannot be updated. More info: https://kubernetes.io/docs/concepts/overview/working-with-objects/names#names +optional
Namespace defines the space within which each name must be unique. An empty namespace is equivalent to the "default" namespace, but "default" is the canonical representation. Not all objects are required to be scoped to a namespace - the value of this field for those objects will be empty.
Must be a DNS_LABEL. Cannot be updated. More info: https://kubernetes.io/docs/concepts/overview/working-with-objects/namespaces +optional
An opaque value that represents the internal version of this object that can be used by clients to determine when objects have changed. May be used for optimistic concurrency, change detection, and the watch operation on a resource or set of resources. Clients must treat these values as opaque and passed unmodified back to the server. They may only be valid for a particular resource or set of resources.
Populated by the system. Read-only. Value must be treated as opaque by clients and . More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#concurrency-control-and-consistency +optional
Deprecated: selfLink is a legacy read-only field that is no longer populated by the system. +optional
UID is a type that holds unique ID values, including UUIDs. Because we don't ONLY use UUIDs, this is an alias to string. Being a type captures intent and helps make sure that UIDs and names do not get conflated.
Seldon Model
ErrorDefault Error
Read the specified Seldon Pipeline Explainer
Name identifies a resource
Namespace provides a logical grouping of resources
Seldon Model
ErrorDefault Error
Read the specified Seldon Pipeline Explainer
Name identifies a resource
Namespace provides a logical grouping of resources
Action
Message
Generic
ErrorDefault Error
No content
Create Seldon Pipeline explanation
Namespace provides a logical grouping of resources
Name identifies a resource
Generic
ErrorDefault Error
No content
Promote the specified Seldon Pipeline Shadow
Name identifies a resource
Namespace provides a logical grouping of resources
Action
Message
Generic
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
No content
Create the specified Seldon Pipeline Shadow experiment
Name identifies a resource
Namespace provides a logical grouping of resources
Action
Message
Generic
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
No content
Delete the specified Seldon Pipeline Shadow
Name identifies a resource
Namespace provides a logical grouping of resources
Action
Message
Generic
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
No content
Promote the specified Seldon Pipeline Canary
Name identifies a resource
Namespace provides a logical grouping of resources
Action
Message
Generic
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
No content
Create the specified Seldon Pipeline Canary experiment
Name identifies a resource
Namespace provides a logical grouping of resources
Action
Message
Weight of the traffic directed to the canary pipeline, between 0 and 100
Generic
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
No content
Delete the specified Seldon Pipeline Canary
Name identifies a resource
Namespace provides a logical grouping of resources
Action
Message
Generic
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
No content
Update the specified Seldon Pipeline Experiment
Name identifies a resource
Namespace provides a logical grouping of resources
Action
Message
Experiment is the Schema for the experiments API
APIVersion defines the versioned schema of this representation of an object. Servers should convert recognized schemas to the latest internal value, and may reject unrecognized values. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#resources +optional
CreationTimestamp is a timestamp representing the server time when this object was created. It is not guaranteed to be set in happens-before order across separate operations. Clients may not set this value. It is represented in RFC3339 form and is in UTC.
Populated by the system. Read-only. Null for lists. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#metadata +optional
Number of seconds allowed for this object to gracefully terminate before it will be removed from the system. Only set when deletionTimestamp is also set. May only be shortened. Read-only. +optional
DeletionTimestamp is RFC 3339 date and time at which this resource will be deleted. This field is set by the server when a graceful deletion is requested by the user, and is not directly settable by a client. The resource is expected to be deleted (no longer visible from resource lists, and not reachable by name) after the time in this field, once the finalizers list is empty. As long as the finalizers list contains items, deletion is blocked. Once the deletionTimestamp is set, this value may not be unset or be set further into the future, although it may be shortened or the resource may be deleted prior to this time. For example, a user may request that a pod is deleted in 30 seconds. The Kubelet will react by sending a graceful termination signal to the containers in the pod. After that 30 seconds, the Kubelet will send a hard termination signal (SIGKILL) to the container and after cleanup, remove the pod from the API. In the presence of network partitions, this object may still exist after this timestamp, until an administrator or automated process can determine the resource is fully terminated. If not set, graceful deletion of the object has not been requested.
Populated by the system when a graceful deletion is requested. Read-only. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#metadata +optional
Must be empty before the object is deleted from the registry. Each entry is an identifier for the responsible component that will remove the entry from the list. If the deletionTimestamp of the object is non-nil, entries in this list can only be removed. Finalizers may be processed and removed in any order. Order is NOT enforced because it introduces significant risk of stuck finalizers. finalizers is a shared field, any actor with permission can reorder it. If the finalizer list is processed in order, then this can lead to a situation in which the component responsible for the first finalizer in the list is waiting for a signal (field value, external system, or other) produced by a component responsible for a finalizer later in the list, resulting in a deadlock. Without enforced ordering finalizers are free to order amongst themselves and are not vulnerable to ordering changes in the list. +optional +patchStrategy=merge
GenerateName is an optional prefix, used by the server, to generate a unique name ONLY IF the Name field has not been provided. If this field is used, the name returned to the client will be different than the name passed. This value will also be combined with a unique suffix. The provided value has the same validation rules as the Name field, and may be truncated by the length of the suffix required to make the value unique on the server.
If this field is specified and the generated name exists, the server will return a 409.
Applied only if Name is not specified. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#idempotency +optional
A sequence number representing a specific generation of the desired state. Populated by the system. Read-only. +optional
Kind is a string value representing the REST resource this object represents. Servers may infer this from the endpoint the client submits requests to. Cannot be updated. In CamelCase. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#types-kinds +optional
Name must be unique within a namespace. Is required when creating resources, although some resources may allow a client to request the generation of an appropriate name automatically. Name is primarily intended for creation idempotence and configuration definition. Cannot be updated. More info: https://kubernetes.io/docs/concepts/overview/working-with-objects/names#names +optional
Namespace defines the space within which each name must be unique. An empty namespace is equivalent to the "default" namespace, but "default" is the canonical representation. Not all objects are required to be scoped to a namespace - the value of this field for those objects will be empty.
Must be a DNS_LABEL. Cannot be updated. More info: https://kubernetes.io/docs/concepts/overview/working-with-objects/namespaces +optional
An opaque value that represents the internal version of this object that can be used by clients to determine when objects have changed. May be used for optimistic concurrency, change detection, and the watch operation on a resource or set of resources. Clients must treat these values as opaque and passed unmodified back to the server. They may only be valid for a particular resource or set of resources.
Populated by the system. Read-only. Value must be treated as opaque by clients and . More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#concurrency-control-and-consistency +optional
Deprecated: selfLink is a legacy read-only field that is no longer populated by the system. +optional
UID is a type that holds unique ID values, including UUIDs. Because we don't ONLY use UUIDs, this is an alias to string. Being a type captures intent and helps make sure that UIDs and names do not get conflated.
Seldon Experiment
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
Create a Seldon Pipeline Experiment
Name identifies a resource
Namespace provides a logical grouping of resources
Action
Message
Experiment is the Schema for the experiments API
APIVersion defines the versioned schema of this representation of an object. Servers should convert recognized schemas to the latest internal value, and may reject unrecognized values. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#resources +optional
CreationTimestamp is a timestamp representing the server time when this object was created. It is not guaranteed to be set in happens-before order across separate operations. Clients may not set this value. It is represented in RFC3339 form and is in UTC.
Populated by the system. Read-only. Null for lists. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#metadata +optional
Number of seconds allowed for this object to gracefully terminate before it will be removed from the system. Only set when deletionTimestamp is also set. May only be shortened. Read-only. +optional
DeletionTimestamp is RFC 3339 date and time at which this resource will be deleted. This field is set by the server when a graceful deletion is requested by the user, and is not directly settable by a client. The resource is expected to be deleted (no longer visible from resource lists, and not reachable by name) after the time in this field, once the finalizers list is empty. As long as the finalizers list contains items, deletion is blocked. Once the deletionTimestamp is set, this value may not be unset or be set further into the future, although it may be shortened or the resource may be deleted prior to this time. For example, a user may request that a pod is deleted in 30 seconds. The Kubelet will react by sending a graceful termination signal to the containers in the pod. After that 30 seconds, the Kubelet will send a hard termination signal (SIGKILL) to the container and after cleanup, remove the pod from the API. In the presence of network partitions, this object may still exist after this timestamp, until an administrator or automated process can determine the resource is fully terminated. If not set, graceful deletion of the object has not been requested.
Populated by the system when a graceful deletion is requested. Read-only. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#metadata +optional
Must be empty before the object is deleted from the registry. Each entry is an identifier for the responsible component that will remove the entry from the list. If the deletionTimestamp of the object is non-nil, entries in this list can only be removed. Finalizers may be processed and removed in any order. Order is NOT enforced because it introduces significant risk of stuck finalizers. finalizers is a shared field, any actor with permission can reorder it. If the finalizer list is processed in order, then this can lead to a situation in which the component responsible for the first finalizer in the list is waiting for a signal (field value, external system, or other) produced by a component responsible for a finalizer later in the list, resulting in a deadlock. Without enforced ordering finalizers are free to order amongst themselves and are not vulnerable to ordering changes in the list. +optional +patchStrategy=merge
GenerateName is an optional prefix, used by the server, to generate a unique name ONLY IF the Name field has not been provided. If this field is used, the name returned to the client will be different than the name passed. This value will also be combined with a unique suffix. The provided value has the same validation rules as the Name field, and may be truncated by the length of the suffix required to make the value unique on the server.
If this field is specified and the generated name exists, the server will return a 409.
Applied only if Name is not specified. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#idempotency +optional
A sequence number representing a specific generation of the desired state. Populated by the system. Read-only. +optional
Kind is a string value representing the REST resource this object represents. Servers may infer this from the endpoint the client submits requests to. Cannot be updated. In CamelCase. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#types-kinds +optional
Name must be unique within a namespace. Is required when creating resources, although some resources may allow a client to request the generation of an appropriate name automatically. Name is primarily intended for creation idempotence and configuration definition. Cannot be updated. More info: https://kubernetes.io/docs/concepts/overview/working-with-objects/names#names +optional
Namespace defines the space within which each name must be unique. An empty namespace is equivalent to the "default" namespace, but "default" is the canonical representation. Not all objects are required to be scoped to a namespace - the value of this field for those objects will be empty.
Must be a DNS_LABEL. Cannot be updated. More info: https://kubernetes.io/docs/concepts/overview/working-with-objects/namespaces +optional
An opaque value that represents the internal version of this object that can be used by clients to determine when objects have changed. May be used for optimistic concurrency, change detection, and the watch operation on a resource or set of resources. Clients must treat these values as opaque and passed unmodified back to the server. They may only be valid for a particular resource or set of resources.
Populated by the system. Read-only. Value must be treated as opaque by clients and . More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#concurrency-control-and-consistency +optional
Deprecated: selfLink is a legacy read-only field that is no longer populated by the system. +optional
UID is a type that holds unique ID values, including UUIDs. Because we don't ONLY use UUIDs, this is an alias to string. Being a type captures intent and helps make sure that UIDs and names do not get conflated.
Seldon Experiment
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
Read Seldon Pipeline Experiment
Name identifies a resource
Namespace provides a logical grouping of resources
Seldon Experiment
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
Delete Seldon Pipeline Experiment Pipelines
Name identifies a resource
Namespace provides a logical grouping of resources
Generic
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
No content
Get details on the pipeline batch job
Name identifies a resource
Namespace provides a logical grouping of resources
JobName identifies a job name
Batch Job Response
ErrorDefault Error
Create the pipeline batch jobs
Name identifies a resource
Namespace provides a logical grouping of resources
BatchJobDefinition is a common interface for BatchJob v1 and v2 which specifies necessary arguments
Batch Data Type (data, json or str)
Interval between batches
Batch Method (predict, feedback (for SCv1))
Batch Payload Type (ndarray, tensor, tftensor, v2raw, v2binary - only if DataType=data)
Number of retries for each instance
Size of the batch (number of predictions per request)
Batch Transport Protocol (rest or grpc)
Number of batch workers
S3 URI of input data file
Container Resources for running batch jobs: limits: CPU
Container Resources for running batch jobs: limits: Memory
name of Kubernetes Secret with S3 credentials
S3 URI of output data file
Size of PVC required for the batch job
Container Resources for running batch jobs: requests: CPU
Container Resources for running batch jobs: requests: Memory
Batch Job Post Response
ErrorDefault Error
ErrorDefault Error
Read the pipeline batch jobs
Name identifies a resource
Namespace provides a logical grouping of resources
Limit of items returned in one response
Requested page
Batch Jobs List Response
ErrorDefault Error
ErrorDefault Error
Update the specified Seldon Pipeline
Name identifies a resource
Namespace provides a logical grouping of resources
Action
Message
Pipeline is the Schema for the pipelines API
APIVersion defines the versioned schema of this representation of an object. Servers should convert recognized schemas to the latest internal value, and may reject unrecognized values. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#resources +optional
CreationTimestamp is a timestamp representing the server time when this object was created. It is not guaranteed to be set in happens-before order across separate operations. Clients may not set this value. It is represented in RFC3339 form and is in UTC.
Populated by the system. Read-only. Null for lists. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#metadata +optional
Number of seconds allowed for this object to gracefully terminate before it will be removed from the system. Only set when deletionTimestamp is also set. May only be shortened. Read-only. +optional
DeletionTimestamp is RFC 3339 date and time at which this resource will be deleted. This field is set by the server when a graceful deletion is requested by the user, and is not directly settable by a client. The resource is expected to be deleted (no longer visible from resource lists, and not reachable by name) after the time in this field, once the finalizers list is empty. As long as the finalizers list contains items, deletion is blocked. Once the deletionTimestamp is set, this value may not be unset or be set further into the future, although it may be shortened or the resource may be deleted prior to this time. For example, a user may request that a pod is deleted in 30 seconds. The Kubelet will react by sending a graceful termination signal to the containers in the pod. After that 30 seconds, the Kubelet will send a hard termination signal (SIGKILL) to the container and after cleanup, remove the pod from the API. In the presence of network partitions, this object may still exist after this timestamp, until an administrator or automated process can determine the resource is fully terminated. If not set, graceful deletion of the object has not been requested.
Populated by the system when a graceful deletion is requested. Read-only. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#metadata +optional
Must be empty before the object is deleted from the registry. Each entry is an identifier for the responsible component that will remove the entry from the list. If the deletionTimestamp of the object is non-nil, entries in this list can only be removed. Finalizers may be processed and removed in any order. Order is NOT enforced because it introduces significant risk of stuck finalizers. finalizers is a shared field, any actor with permission can reorder it. If the finalizer list is processed in order, then this can lead to a situation in which the component responsible for the first finalizer in the list is waiting for a signal (field value, external system, or other) produced by a component responsible for a finalizer later in the list, resulting in a deadlock. Without enforced ordering finalizers are free to order amongst themselves and are not vulnerable to ordering changes in the list. +optional +patchStrategy=merge
GenerateName is an optional prefix, used by the server, to generate a unique name ONLY IF the Name field has not been provided. If this field is used, the name returned to the client will be different than the name passed. This value will also be combined with a unique suffix. The provided value has the same validation rules as the Name field, and may be truncated by the length of the suffix required to make the value unique on the server.
If this field is specified and the generated name exists, the server will return a 409.
Applied only if Name is not specified. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#idempotency +optional
A sequence number representing a specific generation of the desired state. Populated by the system. Read-only. +optional
Kind is a string value representing the REST resource this object represents. Servers may infer this from the endpoint the client submits requests to. Cannot be updated. In CamelCase. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#types-kinds +optional
Name must be unique within a namespace. Is required when creating resources, although some resources may allow a client to request the generation of an appropriate name automatically. Name is primarily intended for creation idempotence and configuration definition. Cannot be updated. More info: https://kubernetes.io/docs/concepts/overview/working-with-objects/names#names +optional
Namespace defines the space within which each name must be unique. An empty namespace is equivalent to the "default" namespace, but "default" is the canonical representation. Not all objects are required to be scoped to a namespace - the value of this field for those objects will be empty.
Must be a DNS_LABEL. Cannot be updated. More info: https://kubernetes.io/docs/concepts/overview/working-with-objects/namespaces +optional
An opaque value that represents the internal version of this object that can be used by clients to determine when objects have changed. May be used for optimistic concurrency, change detection, and the watch operation on a resource or set of resources. Clients must treat these values as opaque and passed unmodified back to the server. They may only be valid for a particular resource or set of resources.
Populated by the system. Read-only. Value must be treated as opaque by clients and . More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#concurrency-control-and-consistency +optional
Deprecated: selfLink is a legacy read-only field that is no longer populated by the system. +optional
UID is a type that holds unique ID values, including UUIDs. Because we don't ONLY use UUIDs, this is an alias to string. Being a type captures intent and helps make sure that UIDs and names do not get conflated.
Seldon Pipeline
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
Read the specified Seldon Pipeline
Name identifies a resource
Namespace provides a logical grouping of resources
Seldon Pipeline
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
Delete the specified Seldon Pipeline
Name identifies a resource
Namespace provides a logical grouping of resources
Action
Message
Generic
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
No content
Create a Seldon Pipeline
Namespace provides a logical grouping of resources
Action
Message
Pipeline is the Schema for the pipelines API
APIVersion defines the versioned schema of this representation of an object. Servers should convert recognized schemas to the latest internal value, and may reject unrecognized values. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#resources +optional
CreationTimestamp is a timestamp representing the server time when this object was created. It is not guaranteed to be set in happens-before order across separate operations. Clients may not set this value. It is represented in RFC3339 form and is in UTC.
Populated by the system. Read-only. Null for lists. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#metadata +optional
Number of seconds allowed for this object to gracefully terminate before it will be removed from the system. Only set when deletionTimestamp is also set. May only be shortened. Read-only. +optional
DeletionTimestamp is RFC 3339 date and time at which this resource will be deleted. This field is set by the server when a graceful deletion is requested by the user, and is not directly settable by a client. The resource is expected to be deleted (no longer visible from resource lists, and not reachable by name) after the time in this field, once the finalizers list is empty. As long as the finalizers list contains items, deletion is blocked. Once the deletionTimestamp is set, this value may not be unset or be set further into the future, although it may be shortened or the resource may be deleted prior to this time. For example, a user may request that a pod is deleted in 30 seconds. The Kubelet will react by sending a graceful termination signal to the containers in the pod. After that 30 seconds, the Kubelet will send a hard termination signal (SIGKILL) to the container and after cleanup, remove the pod from the API. In the presence of network partitions, this object may still exist after this timestamp, until an administrator or automated process can determine the resource is fully terminated. If not set, graceful deletion of the object has not been requested.
Populated by the system when a graceful deletion is requested. Read-only. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#metadata +optional
Must be empty before the object is deleted from the registry. Each entry is an identifier for the responsible component that will remove the entry from the list. If the deletionTimestamp of the object is non-nil, entries in this list can only be removed. Finalizers may be processed and removed in any order. Order is NOT enforced because it introduces significant risk of stuck finalizers. finalizers is a shared field, any actor with permission can reorder it. If the finalizer list is processed in order, then this can lead to a situation in which the component responsible for the first finalizer in the list is waiting for a signal (field value, external system, or other) produced by a component responsible for a finalizer later in the list, resulting in a deadlock. Without enforced ordering finalizers are free to order amongst themselves and are not vulnerable to ordering changes in the list. +optional +patchStrategy=merge
GenerateName is an optional prefix, used by the server, to generate a unique name ONLY IF the Name field has not been provided. If this field is used, the name returned to the client will be different than the name passed. This value will also be combined with a unique suffix. The provided value has the same validation rules as the Name field, and may be truncated by the length of the suffix required to make the value unique on the server.
If this field is specified and the generated name exists, the server will return a 409.
Applied only if Name is not specified. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#idempotency +optional
A sequence number representing a specific generation of the desired state. Populated by the system. Read-only. +optional
Kind is a string value representing the REST resource this object represents. Servers may infer this from the endpoint the client submits requests to. Cannot be updated. In CamelCase. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#types-kinds +optional
Name must be unique within a namespace. Is required when creating resources, although some resources may allow a client to request the generation of an appropriate name automatically. Name is primarily intended for creation idempotence and configuration definition. Cannot be updated. More info: https://kubernetes.io/docs/concepts/overview/working-with-objects/names#names +optional
Namespace defines the space within which each name must be unique. An empty namespace is equivalent to the "default" namespace, but "default" is the canonical representation. Not all objects are required to be scoped to a namespace - the value of this field for those objects will be empty.
Must be a DNS_LABEL. Cannot be updated. More info: https://kubernetes.io/docs/concepts/overview/working-with-objects/namespaces +optional
An opaque value that represents the internal version of this object that can be used by clients to determine when objects have changed. May be used for optimistic concurrency, change detection, and the watch operation on a resource or set of resources. Clients must treat these values as opaque and passed unmodified back to the server. They may only be valid for a particular resource or set of resources.
Populated by the system. Read-only. Value must be treated as opaque by clients and . More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#concurrency-control-and-consistency +optional
Deprecated: selfLink is a legacy read-only field that is no longer populated by the system. +optional
UID is a type that holds unique ID values, including UUIDs. Because we don't ONLY use UUIDs, this is an alias to string. Being a type captures intent and helps make sure that UIDs and names do not get conflated.
Seldon Pipeline
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
list objects of kind Seldon Pipelines
Namespace provides a logical grouping of resources
Seldon Pipeline List
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
Read the git commits for a model
Name identifies a resource
Namespace provides a logical grouping of resources
Git Logs
ErrorDefault Error
Read the git diff for a model
Name identifies a resource
Namespace provides a logical grouping of resources
Hash
Git Diff
ErrorDefault Error
Update the specified Seldon Model
Name identifies a resource
Namespace provides a logical grouping of resources
Action
Message
Model is the Schema for the models API
APIVersion defines the versioned schema of this representation of an object. Servers should convert recognized schemas to the latest internal value, and may reject unrecognized values. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#resources +optional
CreationTimestamp is a timestamp representing the server time when this object was created. It is not guaranteed to be set in happens-before order across separate operations. Clients may not set this value. It is represented in RFC3339 form and is in UTC.
Populated by the system. Read-only. Null for lists. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#metadata +optional
Number of seconds allowed for this object to gracefully terminate before it will be removed from the system. Only set when deletionTimestamp is also set. May only be shortened. Read-only. +optional
DeletionTimestamp is RFC 3339 date and time at which this resource will be deleted. This field is set by the server when a graceful deletion is requested by the user, and is not directly settable by a client. The resource is expected to be deleted (no longer visible from resource lists, and not reachable by name) after the time in this field, once the finalizers list is empty. As long as the finalizers list contains items, deletion is blocked. Once the deletionTimestamp is set, this value may not be unset or be set further into the future, although it may be shortened or the resource may be deleted prior to this time. For example, a user may request that a pod is deleted in 30 seconds. The Kubelet will react by sending a graceful termination signal to the containers in the pod. After that 30 seconds, the Kubelet will send a hard termination signal (SIGKILL) to the container and after cleanup, remove the pod from the API. In the presence of network partitions, this object may still exist after this timestamp, until an administrator or automated process can determine the resource is fully terminated. If not set, graceful deletion of the object has not been requested.
Populated by the system when a graceful deletion is requested. Read-only. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#metadata +optional
Must be empty before the object is deleted from the registry. Each entry is an identifier for the responsible component that will remove the entry from the list. If the deletionTimestamp of the object is non-nil, entries in this list can only be removed. Finalizers may be processed and removed in any order. Order is NOT enforced because it introduces significant risk of stuck finalizers. finalizers is a shared field, any actor with permission can reorder it. If the finalizer list is processed in order, then this can lead to a situation in which the component responsible for the first finalizer in the list is waiting for a signal (field value, external system, or other) produced by a component responsible for a finalizer later in the list, resulting in a deadlock. Without enforced ordering finalizers are free to order amongst themselves and are not vulnerable to ordering changes in the list. +optional +patchStrategy=merge
GenerateName is an optional prefix, used by the server, to generate a unique name ONLY IF the Name field has not been provided. If this field is used, the name returned to the client will be different than the name passed. This value will also be combined with a unique suffix. The provided value has the same validation rules as the Name field, and may be truncated by the length of the suffix required to make the value unique on the server.
If this field is specified and the generated name exists, the server will return a 409.
Applied only if Name is not specified. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#idempotency +optional
A sequence number representing a specific generation of the desired state. Populated by the system. Read-only. +optional
Kind is a string value representing the REST resource this object represents. Servers may infer this from the endpoint the client submits requests to. Cannot be updated. In CamelCase. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#types-kinds +optional
Name must be unique within a namespace. Is required when creating resources, although some resources may allow a client to request the generation of an appropriate name automatically. Name is primarily intended for creation idempotence and configuration definition. Cannot be updated. More info: https://kubernetes.io/docs/concepts/overview/working-with-objects/names#names +optional
Namespace defines the space within which each name must be unique. An empty namespace is equivalent to the "default" namespace, but "default" is the canonical representation. Not all objects are required to be scoped to a namespace - the value of this field for those objects will be empty.
Must be a DNS_LABEL. Cannot be updated. More info: https://kubernetes.io/docs/concepts/overview/working-with-objects/namespaces +optional
An opaque value that represents the internal version of this object that can be used by clients to determine when objects have changed. May be used for optimistic concurrency, change detection, and the watch operation on a resource or set of resources. Clients must treat these values as opaque and passed unmodified back to the server. They may only be valid for a particular resource or set of resources.
Populated by the system. Read-only. Value must be treated as opaque by clients and . More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#concurrency-control-and-consistency +optional
Deprecated: selfLink is a legacy read-only field that is no longer populated by the system. +optional
UID is a type that holds unique ID values, including UUIDs. Because we don't ONLY use UUIDs, this is an alias to string. Being a type captures intent and helps make sure that UIDs and names do not get conflated.
Seldon Model
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
Read the specified Seldon Model
Name identifies a resource
Namespace provides a logical grouping of resources
Seldon Model
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
Delete the specified Seldon Model
Name identifies a resource
Namespace provides a logical grouping of resources
Action
Message
Generic
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
No content
Create a Seldon Model
Namespace provides a logical grouping of resources
Action
Message
Model is the Schema for the models API
APIVersion defines the versioned schema of this representation of an object. Servers should convert recognized schemas to the latest internal value, and may reject unrecognized values. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#resources +optional
CreationTimestamp is a timestamp representing the server time when this object was created. It is not guaranteed to be set in happens-before order across separate operations. Clients may not set this value. It is represented in RFC3339 form and is in UTC.
Populated by the system. Read-only. Null for lists. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#metadata +optional
Number of seconds allowed for this object to gracefully terminate before it will be removed from the system. Only set when deletionTimestamp is also set. May only be shortened. Read-only. +optional
DeletionTimestamp is RFC 3339 date and time at which this resource will be deleted. This field is set by the server when a graceful deletion is requested by the user, and is not directly settable by a client. The resource is expected to be deleted (no longer visible from resource lists, and not reachable by name) after the time in this field, once the finalizers list is empty. As long as the finalizers list contains items, deletion is blocked. Once the deletionTimestamp is set, this value may not be unset or be set further into the future, although it may be shortened or the resource may be deleted prior to this time. For example, a user may request that a pod is deleted in 30 seconds. The Kubelet will react by sending a graceful termination signal to the containers in the pod. After that 30 seconds, the Kubelet will send a hard termination signal (SIGKILL) to the container and after cleanup, remove the pod from the API. In the presence of network partitions, this object may still exist after this timestamp, until an administrator or automated process can determine the resource is fully terminated. If not set, graceful deletion of the object has not been requested.
Populated by the system when a graceful deletion is requested. Read-only. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#metadata +optional
Must be empty before the object is deleted from the registry. Each entry is an identifier for the responsible component that will remove the entry from the list. If the deletionTimestamp of the object is non-nil, entries in this list can only be removed. Finalizers may be processed and removed in any order. Order is NOT enforced because it introduces significant risk of stuck finalizers. finalizers is a shared field, any actor with permission can reorder it. If the finalizer list is processed in order, then this can lead to a situation in which the component responsible for the first finalizer in the list is waiting for a signal (field value, external system, or other) produced by a component responsible for a finalizer later in the list, resulting in a deadlock. Without enforced ordering finalizers are free to order amongst themselves and are not vulnerable to ordering changes in the list. +optional +patchStrategy=merge
GenerateName is an optional prefix, used by the server, to generate a unique name ONLY IF the Name field has not been provided. If this field is used, the name returned to the client will be different than the name passed. This value will also be combined with a unique suffix. The provided value has the same validation rules as the Name field, and may be truncated by the length of the suffix required to make the value unique on the server.
If this field is specified and the generated name exists, the server will return a 409.
Applied only if Name is not specified. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#idempotency +optional
A sequence number representing a specific generation of the desired state. Populated by the system. Read-only. +optional
Kind is a string value representing the REST resource this object represents. Servers may infer this from the endpoint the client submits requests to. Cannot be updated. In CamelCase. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#types-kinds +optional
Name must be unique within a namespace. Is required when creating resources, although some resources may allow a client to request the generation of an appropriate name automatically. Name is primarily intended for creation idempotence and configuration definition. Cannot be updated. More info: https://kubernetes.io/docs/concepts/overview/working-with-objects/names#names +optional
Namespace defines the space within which each name must be unique. An empty namespace is equivalent to the "default" namespace, but "default" is the canonical representation. Not all objects are required to be scoped to a namespace - the value of this field for those objects will be empty.
Must be a DNS_LABEL. Cannot be updated. More info: https://kubernetes.io/docs/concepts/overview/working-with-objects/namespaces +optional
An opaque value that represents the internal version of this object that can be used by clients to determine when objects have changed. May be used for optimistic concurrency, change detection, and the watch operation on a resource or set of resources. Clients must treat these values as opaque and passed unmodified back to the server. They may only be valid for a particular resource or set of resources.
Populated by the system. Read-only. Value must be treated as opaque by clients and . More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#concurrency-control-and-consistency +optional
Deprecated: selfLink is a legacy read-only field that is no longer populated by the system. +optional
UID is a type that holds unique ID values, including UUIDs. Because we don't ONLY use UUIDs, this is an alias to string. Being a type captures intent and helps make sure that UIDs and names do not get conflated.
Seldon Model
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
ErrorDefault Error
Read the GitOps status
Namespace provides a logical grouping of resources
Generic
Generic
No content
Read the git diffs from hashes
Namespace provides a logical grouping of resources
Hashes is an array of git commit hashes to get diffs for
Git Diffs
ErrorDefault Error
Last updated
Was this helpful?