Class GkeInferenceQuickstartClient (0.1.0) 

GkeInferenceQuickstartClient(
    *,
    credentials: typing.Optional[google.auth.credentials.Credentials] = None,
    transport: typing.Optional[
        typing.Union[
            str,
            google.cloud.gkerecommender_v1.services.gke_inference_quickstart.transports.base.GkeInferenceQuickstartTransport,
            typing.Callable[
                [...],
                google.cloud.gkerecommender_v1.services.gke_inference_quickstart.transports.base.GkeInferenceQuickstartTransport,
            ],
        ]
    ] = None,
    client_options: typing.Optional[
        typing.Union[google.api_core.client_options.ClientOptions, dict]
    ] = None,
    client_info: google.api_core.gapic_v1.client_info.ClientInfo = google.api_core.gapic_v1.client_info.ClientInfo
)

GKE Inference Quickstart (GIQ) service provides profiles with performance metrics for popular models and model servers across multiple accelerators. These profiles help generate optimized best practices for running inference on GKE.

Properties

api_endpoint

Return the API endpoint used by the client instance.

Returns
Type Description
str The API endpoint used by the client instance.

transport

Returns the transport used by the client instance.

Returns
Type Description
GkeInferenceQuickstartTransport The transport used by the client instance.

universe_domain

Return the universe domain used by the client instance.

Returns
Type Description
str The universe domain used by the client instance.

Methods

GkeInferenceQuickstartClient

GkeInferenceQuickstartClient(
    *,
    credentials: typing.Optional[google.auth.credentials.Credentials] = None,
    transport: typing.Optional[
        typing.Union[
            str,
            google.cloud.gkerecommender_v1.services.gke_inference_quickstart.transports.base.GkeInferenceQuickstartTransport,
            typing.Callable[
                [...],
                google.cloud.gkerecommender_v1.services.gke_inference_quickstart.transports.base.GkeInferenceQuickstartTransport,
            ],
        ]
    ] = None,
    client_options: typing.Optional[
        typing.Union[google.api_core.client_options.ClientOptions, dict]
    ] = None,
    client_info: google.api_core.gapic_v1.client_info.ClientInfo = google.api_core.gapic_v1.client_info.ClientInfo
)

Instantiates the gke inference quickstart client.

Parameters
Name Description
credentials Optional[google.auth.credentials.Credentials]

The authorization credentials to attach to requests. These credentials identify the application to the service; if none are specified, the client will attempt to ascertain the credentials from the environment.
transport Optional[Union[str,GkeInferenceQuickstartTransport,Callable[..., GkeInferenceQuickstartTransport]]]

The transport to use, or a Callable that constructs and returns a new transport. If a Callable is given, it will be called with the same set of initialization arguments as used in the GkeInferenceQuickstartTransport constructor. If set to None, a transport is chosen automatically.
client_options Optional[Union[google.api_core.client_options.ClientOptions, dict]]

Custom options for the client. 1. The api_endpoint property can be used to override the default endpoint provided by the client when transport is not explicitly provided. Only if this property is not set and transport was not explicitly provided, the endpoint is determined by the GOOGLE_API_USE_MTLS_ENDPOINT environment variable, which have one of the following values: "always" (always use the default mTLS endpoint), "never" (always use the default regular endpoint) and "auto" (auto-switch to the default mTLS endpoint if client certificate is present; this is the default value). 2. If the GOOGLE_API_USE_CLIENT_CERTIFICATE environment variable is "true", then the client_cert_source property can be used to provide a client certificate for mTLS transport. If not provided, the default SSL client certificate will be used if present. If GOOGLE_API_USE_CLIENT_CERTIFICATE is "false" or not set, no client certificate will be used. 3. The universe_domain property can be used to override the default "googleapis.com" universe. Note that the api_endpoint property still takes precedence; and universe_domain is currently not supported for mTLS.
client_info google.api_core.gapic_v1.client_info.ClientInfo

The client info used to send a user-agent string along with API requests. If None, then default info will be used. Generally, you only need to set this if you're developing your own client library.
Exceptions
Type Description
google.auth.exceptions.MutualTLSChannelError If mutual TLS transport creation failed for any reason.

__exit__

__exit__(type, value, traceback)

Releases underlying transport's resources.

common_billing_account_path

common_billing_account_path(billing_account: str) -> str

Returns a fully-qualified billing_account string.

common_folder_path

common_folder_path(folder: str) -> str

Returns a fully-qualified folder string.

common_location_path

common_location_path(project: str, location: str) -> str

Returns a fully-qualified location string.

common_organization_path

common_organization_path(organization: str) -> str

Returns a fully-qualified organization string.

common_project_path

common_project_path(project: str) -> str

Returns a fully-qualified project string.

fetch_benchmarking_data

fetch_benchmarking_data(
    request: typing.Optional[
        typing.Union[
            google.cloud.gkerecommender_v1.types.gkerecommender.FetchBenchmarkingDataRequest,
            dict,
        ]
    ] = None,
    *,
    retry: typing.Optional[
        typing.Union[
            google.api_core.retry.retry_unary.Retry,
            google.api_core.gapic_v1.method._MethodDefault,
        ]
    ] = _MethodDefault._DEFAULT_VALUE,
    timeout: typing.Union[float, object] = _MethodDefault._DEFAULT_VALUE,
    metadata: typing.Sequence[typing.Tuple[str, typing.Union[str, bytes]]] = ()
) -> google.cloud.gkerecommender_v1.types.gkerecommender.FetchBenchmarkingDataResponse

Fetches all of the benchmarking data available for a profile. Benchmarking data returns all of the performance metrics available for a given model server setup on a given instance type.

# This snippet has been automatically generated and should be regarded as a
# code template only.
# It will require modifications to work:
# - It may require correct/in-range values for request initialization.
# - It may require specifying regional endpoints when creating the service
#   client as shown in:
#   https://googleapis.dev/python/google-api-core/latest/client_options.html
from google.cloud import gkerecommender_v1

def sample_fetch_benchmarking_data():
    # Create a client
    client = gkerecommender_v1.GkeInferenceQuickstartClient()

    # Initialize request argument(s)
    model_server_info = gkerecommender_v1.ModelServerInfo()
    model_server_info.model = "model_value"
    model_server_info.model_server = "model_server_value"

    request = gkerecommender_v1.FetchBenchmarkingDataRequest(
        model_server_info=model_server_info,
    )

    # Make the request
    response = client.fetch_benchmarking_data(request=request)

    # Handle the response
    print(response)
Parameters
Name Description
request Union[google.cloud.gkerecommender_v1.types.FetchBenchmarkingDataRequest, dict]

The request object. Request message for GkeInferenceQuickstart.FetchBenchmarkingData.
retry google.api_core.retry.Retry

Designation of what errors, if any, should be retried.
timeout float

The timeout for this request.
metadata Sequence[Tuple[str, Union[str, bytes]]]

Key/value pairs which should be sent along with the request as metadata. Normally, each value must be of type str, but for metadata keys ending with the suffix -bin, the corresponding values must be of type bytes.
Returns
Type Description
google.cloud.gkerecommender_v1.types.FetchBenchmarkingDataResponse Response message for GkeInferenceQuickstart.FetchBenchmarkingData.

fetch_model_server_versions

fetch_model_server_versions(
    request: typing.Optional[
        typing.Union[
            google.cloud.gkerecommender_v1.types.gkerecommender.FetchModelServerVersionsRequest,
            dict,
        ]
    ] = None,
    *,
    retry: typing.Optional[
        typing.Union[
            google.api_core.retry.retry_unary.Retry,
            google.api_core.gapic_v1.method._MethodDefault,
        ]
    ] = _MethodDefault._DEFAULT_VALUE,
    timeout: typing.Union[float, object] = _MethodDefault._DEFAULT_VALUE,
    metadata: typing.Sequence[typing.Tuple[str, typing.Union[str, bytes]]] = ()
) -> (
    google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelServerVersionsPager
)

Fetches available model server versions. Open-source servers use their own versioning schemas (e.g., vllm uses semver like v1.0.0).

Some model servers have different versioning schemas depending on the accelerator. For example, vllm uses semver on GPUs, but returns nightly build tags on TPUs. All available versions will be returned when different schemas are present.

# This snippet has been automatically generated and should be regarded as a
# code template only.
# It will require modifications to work:
# - It may require correct/in-range values for request initialization.
# - It may require specifying regional endpoints when creating the service
#   client as shown in:
#   https://googleapis.dev/python/google-api-core/latest/client_options.html
from google.cloud import gkerecommender_v1

def sample_fetch_model_server_versions():
    # Create a client
    client = gkerecommender_v1.GkeInferenceQuickstartClient()

    # Initialize request argument(s)
    request = gkerecommender_v1.FetchModelServerVersionsRequest(
        model="model_value",
        model_server="model_server_value",
    )

    # Make the request
    page_result = client.fetch_model_server_versions(request=request)

    # Handle the response
    for response in page_result:
        print(response)
Parameters
Name Description
request Union[google.cloud.gkerecommender_v1.types.FetchModelServerVersionsRequest, dict]

The request object. Request message for GkeInferenceQuickstart.FetchModelServerVersions.
retry google.api_core.retry.Retry

Designation of what errors, if any, should be retried.
timeout float

The timeout for this request.
metadata Sequence[Tuple[str, Union[str, bytes]]]

Key/value pairs which should be sent along with the request as metadata. Normally, each value must be of type str, but for metadata keys ending with the suffix -bin, the corresponding values must be of type bytes.
Returns
Type Description
google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelServerVersionsPager Response message for GkeInferenceQuickstart.FetchModelServerVersions. Iterating over this object will yield results and resolve additional pages automatically.

fetch_model_servers

fetch_model_servers(
    request: typing.Optional[
        typing.Union[
            google.cloud.gkerecommender_v1.types.gkerecommender.FetchModelServersRequest,
            dict,
        ]
    ] = None,
    *,
    retry: typing.Optional[
        typing.Union[
            google.api_core.retry.retry_unary.Retry,
            google.api_core.gapic_v1.method._MethodDefault,
        ]
    ] = _MethodDefault._DEFAULT_VALUE,
    timeout: typing.Union[float, object] = _MethodDefault._DEFAULT_VALUE,
    metadata: typing.Sequence[typing.Tuple[str, typing.Union[str, bytes]]] = ()
) -> (
    google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelServersPager
)

Fetches available model servers. Open-source model servers use simplified, lowercase names (e.g., vllm).

# This snippet has been automatically generated and should be regarded as a
# code template only.
# It will require modifications to work:
# - It may require correct/in-range values for request initialization.
# - It may require specifying regional endpoints when creating the service
#   client as shown in:
#   https://googleapis.dev/python/google-api-core/latest/client_options.html
from google.cloud import gkerecommender_v1

def sample_fetch_model_servers():
    # Create a client
    client = gkerecommender_v1.GkeInferenceQuickstartClient()

    # Initialize request argument(s)
    request = gkerecommender_v1.FetchModelServersRequest(
        model="model_value",
    )

    # Make the request
    page_result = client.fetch_model_servers(request=request)

    # Handle the response
    for response in page_result:
        print(response)
Parameters
Name Description
request Union[google.cloud.gkerecommender_v1.types.FetchModelServersRequest, dict]

The request object. Request message for GkeInferenceQuickstart.FetchModelServers.
retry google.api_core.retry.Retry

Designation of what errors, if any, should be retried.
timeout float

The timeout for this request.
metadata Sequence[Tuple[str, Union[str, bytes]]]

Key/value pairs which should be sent along with the request as metadata. Normally, each value must be of type str, but for metadata keys ending with the suffix -bin, the corresponding values must be of type bytes.
Returns
Type Description
google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelServersPager Response message for GkeInferenceQuickstart.FetchModelServers. Iterating over this object will yield results and resolve additional pages automatically.

fetch_models

fetch_models(
    request: typing.Optional[
        typing.Union[
            google.cloud.gkerecommender_v1.types.gkerecommender.FetchModelsRequest, dict
        ]
    ] = None,
    *,
    retry: typing.Optional[
        typing.Union[
            google.api_core.retry.retry_unary.Retry,
            google.api_core.gapic_v1.method._MethodDefault,
        ]
    ] = _MethodDefault._DEFAULT_VALUE,
    timeout: typing.Union[float, object] = _MethodDefault._DEFAULT_VALUE,
    metadata: typing.Sequence[typing.Tuple[str, typing.Union[str, bytes]]] = ()
) -> (
    google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelsPager
)

Fetches available models. Open-source models follow the Huggingface Hub owner/model_name format.

# This snippet has been automatically generated and should be regarded as a
# code template only.
# It will require modifications to work:
# - It may require correct/in-range values for request initialization.
# - It may require specifying regional endpoints when creating the service
#   client as shown in:
#   https://googleapis.dev/python/google-api-core/latest/client_options.html
from google.cloud import gkerecommender_v1

def sample_fetch_models():
    # Create a client
    client = gkerecommender_v1.GkeInferenceQuickstartClient()

    # Initialize request argument(s)
    request = gkerecommender_v1.FetchModelsRequest(
    )

    # Make the request
    page_result = client.fetch_models(request=request)

    # Handle the response
    for response in page_result:
        print(response)
Parameters
Name Description
request Union[google.cloud.gkerecommender_v1.types.FetchModelsRequest, dict]

The request object. Request message for GkeInferenceQuickstart.FetchModels.
retry google.api_core.retry.Retry

Designation of what errors, if any, should be retried.
timeout float

The timeout for this request.
metadata Sequence[Tuple[str, Union[str, bytes]]]

Key/value pairs which should be sent along with the request as metadata. Normally, each value must be of type str, but for metadata keys ending with the suffix -bin, the corresponding values must be of type bytes.
Returns
Type Description
google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchModelsPager Response message for GkeInferenceQuickstart.FetchModels. Iterating over this object will yield results and resolve additional pages automatically.

fetch_profiles

fetch_profiles(
    request: typing.Optional[
        typing.Union[
            google.cloud.gkerecommender_v1.types.gkerecommender.FetchProfilesRequest,
            dict,
        ]
    ] = None,
    *,
    retry: typing.Optional[
        typing.Union[
            google.api_core.retry.retry_unary.Retry,
            google.api_core.gapic_v1.method._MethodDefault,
        ]
    ] = _MethodDefault._DEFAULT_VALUE,
    timeout: typing.Union[float, object] = _MethodDefault._DEFAULT_VALUE,
    metadata: typing.Sequence[typing.Tuple[str, typing.Union[str, bytes]]] = ()
) -> (
    google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchProfilesPager
)

Fetches available profiles. A profile contains performance metrics and cost information for a specific model server setup. Profiles can be filtered by parameters. If no filters are provided, all profiles are returned.

Profiles display a single value per performance metric based on the provided performance requirements. If no requirements are given, the metrics represent the inflection point. See Run best practice inference with GKE Inference Quickstart recipes <https://cloud.google.com/kubernetes-engine/docs/how-to/machine-learning/inference/inference-quickstart#how>__ for details.

# This snippet has been automatically generated and should be regarded as a
# code template only.
# It will require modifications to work:
# - It may require correct/in-range values for request initialization.
# - It may require specifying regional endpoints when creating the service
#   client as shown in:
#   https://googleapis.dev/python/google-api-core/latest/client_options.html
from google.cloud import gkerecommender_v1

def sample_fetch_profiles():
    # Create a client
    client = gkerecommender_v1.GkeInferenceQuickstartClient()

    # Initialize request argument(s)
    request = gkerecommender_v1.FetchProfilesRequest(
    )

    # Make the request
    page_result = client.fetch_profiles(request=request)

    # Handle the response
    for response in page_result:
        print(response)
Parameters
Name Description
request Union[google.cloud.gkerecommender_v1.types.FetchProfilesRequest, dict]

The request object. Request message for GkeInferenceQuickstart.FetchProfiles.
retry google.api_core.retry.Retry

Designation of what errors, if any, should be retried.
timeout float

The timeout for this request.
metadata Sequence[Tuple[str, Union[str, bytes]]]

Key/value pairs which should be sent along with the request as metadata. Normally, each value must be of type str, but for metadata keys ending with the suffix -bin, the corresponding values must be of type bytes.
Returns
Type Description
google.cloud.gkerecommender_v1.services.gke_inference_quickstart.pagers.FetchProfilesPager Response message for GkeInferenceQuickstart.FetchProfiles. Iterating over this object will yield results and resolve additional pages automatically.

from_service_account_file

from_service_account_file(filename: str, *args, **kwargs)

Creates an instance of this client using the provided credentials file.

Parameter
Name Description
filename str

The path to the service account private key json file.
Returns
Type Description
GkeInferenceQuickstartClient The constructed client.

from_service_account_info

from_service_account_info(info: dict, *args, **kwargs)

Creates an instance of this client using the provided credentials info.

Parameter
Name Description
info dict

The service account private key info.
Returns
Type Description
GkeInferenceQuickstartClient The constructed client.

from_service_account_json

from_service_account_json(filename: str, *args, **kwargs)

Creates an instance of this client using the provided credentials file.

Parameter
Name Description
filename str

The path to the service account private key json file.
Returns
Type Description
GkeInferenceQuickstartClient The constructed client.

generate_optimized_manifest

generate_optimized_manifest(
    request: typing.Optional[
        typing.Union[
            google.cloud.gkerecommender_v1.types.gkerecommender.GenerateOptimizedManifestRequest,
            dict,
        ]
    ] = None,
    *,
    retry: typing.Optional[
        typing.Union[
            google.api_core.retry.retry_unary.Retry,
            google.api_core.gapic_v1.method._MethodDefault,
        ]
    ] = _MethodDefault._DEFAULT_VALUE,
    timeout: typing.Union[float, object] = _MethodDefault._DEFAULT_VALUE,
    metadata: typing.Sequence[typing.Tuple[str, typing.Union[str, bytes]]] = ()
) -> (
    google.cloud.gkerecommender_v1.types.gkerecommender.GenerateOptimizedManifestResponse
)

Generates an optimized deployment manifest for a given model and model server, based on the specified accelerator, performance targets, and configurations. See Run best practice inference with GKE Inference Quickstart recipes <https://cloud.google.com/kubernetes-engine/docs/how-to/machine-learning/inference/inference-quickstart>__ for deployment details.

# This snippet has been automatically generated and should be regarded as a
# code template only.
# It will require modifications to work:
# - It may require correct/in-range values for request initialization.
# - It may require specifying regional endpoints when creating the service
#   client as shown in:
#   https://googleapis.dev/python/google-api-core/latest/client_options.html
from google.cloud import gkerecommender_v1

def sample_generate_optimized_manifest():
    # Create a client
    client = gkerecommender_v1.GkeInferenceQuickstartClient()

    # Initialize request argument(s)
    model_server_info = gkerecommender_v1.ModelServerInfo()
    model_server_info.model = "model_value"
    model_server_info.model_server = "model_server_value"

    request = gkerecommender_v1.GenerateOptimizedManifestRequest(
        model_server_info=model_server_info,
        accelerator_type="accelerator_type_value",
    )

    # Make the request
    response = client.generate_optimized_manifest(request=request)

    # Handle the response
    print(response)
Parameters
Name Description
request Union[google.cloud.gkerecommender_v1.types.GenerateOptimizedManifestRequest, dict]

The request object. Request message for GkeInferenceQuickstart.GenerateOptimizedManifest.
retry google.api_core.retry.Retry

Designation of what errors, if any, should be retried.
timeout float

The timeout for this request.
metadata Sequence[Tuple[str, Union[str, bytes]]]

Key/value pairs which should be sent along with the request as metadata. Normally, each value must be of type str, but for metadata keys ending with the suffix -bin, the corresponding values must be of type bytes.
Returns
Type Description
google.cloud.gkerecommender_v1.types.GenerateOptimizedManifestResponse Response message for GkeInferenceQuickstart.GenerateOptimizedManifest.

get_mtls_endpoint_and_cert_source

get_mtls_endpoint_and_cert_source(
    client_options: typing.Optional[
        google.api_core.client_options.ClientOptions
    ] = None,
)

Deprecated. Return the API endpoint and client cert source for mutual TLS.

The client cert source is determined in the following order: (1) if GOOGLE_API_USE_CLIENT_CERTIFICATE environment variable is not "true", the client cert source is None. (2) if client_options.client_cert_source is provided, use the provided one; if the default client cert source exists, use the default one; otherwise the client cert source is None.

The API endpoint is determined in the following order: (1) if client_options.api_endpoint if provided, use the provided one. (2) if GOOGLE_API_USE_CLIENT_CERTIFICATE environment variable is "always", use the default mTLS endpoint; if the environment variable is "never", use the default API endpoint; otherwise if client cert source exists, use the default mTLS endpoint, otherwise use the default API endpoint.

More details can be found at https://google.aip.dev/auth/4114.

Parameter
Name Description
client_options google.api_core.client_options.ClientOptions

Custom options for the client. Only the api_endpoint and client_cert_source properties may be used in this method.
Exceptions
Type Description
google.auth.exceptions.MutualTLSChannelError If any errors happen.
Returns
Type Description
Tuple[str, Callable[[], Tuple[bytes, bytes]]] returns the API endpoint and the client cert source to use.

parse_common_billing_account_path

parse_common_billing_account_path(path: str) -> typing.Dict[str, str]

Parse a billing_account path into its component segments.

parse_common_folder_path

parse_common_folder_path(path: str) -> typing.Dict[str, str]

Parse a folder path into its component segments.

parse_common_location_path

parse_common_location_path(path: str) -> typing.Dict[str, str]

Parse a location path into its component segments.

parse_common_organization_path

parse_common_organization_path(path: str) -> typing.Dict[str, str]

Parse a organization path into its component segments.

parse_common_project_path

parse_common_project_path(path: str) -> typing.Dict[str, str]

Parse a project path into its component segments.