gen_ai_hub.proxy.gen_ai_hub_proxy package

class GenAIHubProxyClient

Bases: BaseProxyClient

GenAIHubProxyClient is a proxy client for interacting with the GenAI Hub.

classmethod add_foundation_model_scenario(scenario_id, config_names=None, prediction_url_suffix=None, model_name_parameter='model_name')

Add a foundational model scenario to the client.

Parameters:

scenario_id (str) -- the scenario ID.
config_names (Optional[List[str]], optional) -- list of configuration names, defaults to None
prediction_url_suffix (Optional[str], optional) -- prediction URL suffix, defaults to None
model_name_parameter (str, optional) -- model name parameter, defaults to 'model_name'

classmethod for_profile(profile=None)

Create a GenAIHubProxyClient instance for the given profile.

Parameters:: profile (str, optional) -- Profile name, defaults to None
Returns:: GenAIHubProxyClient instance.
Return type:: GenAIHubProxyClient

classmethod init_client(data)

Initialize the client with the provided data.

Parameters:: data (Any) -- Input data for client initialization.
Returns:: Initialized data.
Return type:: Any

classmethod set_default_values(**kwargs): Set default values for the client.

get_additional_headers()

Get only the additional headers (instance-level and request-level).

Returns:: Additional headers.
Return type:: Dict[str, str]

get_ai_core_token()

Get the AI core token for authentication.

Returns:: AI core token.
Return type:: str

get_deployments()

Get the list of deployments.

Returns:: List of deployments.
Return type:: List[Deployment]

get_request_header()

Get the request headers for requests made by the client.

Returns:: Request headers.
Return type:: Dict[str, str]

model_post_init(context, /)

This function is meant to behave like a BaseModel method to initialise private attributes.

It takes context as an argument since that's what pydantic-core passes when calling it.

Args:: self: The BaseModel instance. context: The context.

Parameters:

self (BaseModel)
context (Any)

Return type:

None

select_deployment(raise_on_multiple=False, **search_key_value)

Parameters:: raise_on_multiple (bool)

set_headers_addition(headers)

Set additional headers for requests made by the client.

Parameters:: headers (Dict[str, str]) -- Headers to add.

update_deployments()

Update the list of deployments from the GenAI Hub.

Returns:: List of updated deployments.
Return type:: List[Deployment]

AI_CLIENT_TYPE_VAL: ClassVar[str] = 'GenAI Hub SDK (Python)'

ai_core_client: AICoreV2Client | None

auth_url: str | None

base_url: str | None

client_id: str | None

client_secret: str | None

default_values: ClassVar[Dict[str, Any]] = {}

property deployment_class: Type[Deployment]

property deployments: List[Deployment]

foundational_model_scenarios: ClassVar[List[FoundationalModelScenario]] = [FoundationalModelScenario(scenario_id='foundation-models', config_names=['*'], model_name_parameter='model_name', prediction_url_suffix=None)]

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True, 'extra': 'allow', 'protected_namespaces': ()}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

on_invalid_deployments: ClassVar[InvalidDeploymentBehavior] = 'warn'

property request_header: Dict[str, Any]

resource_group: str | None

temporary_headers_addition(headers)

Context manager to temporarily add headers to requests made by the GenAIHubProxyClient.

Parameters:: headers (Dict[str, str]) -- Headers to add temporarily.

Submodules

gen_ai_hub.proxy.gen_ai_hub_proxy.client module

class Deployment

Bases: BaseDeployment

Deployment class represents a deployment of a foundational model in the GenAI Hub.

classmethod get_model_identification_kwargs()

Get model identification keywords.

Returns:: Tuple of model identification keywords.
Return type:: Tuple[str]

additional_request_body_kwargs()

Return type:: Dict[str, Any]

additonal_parameters: Dict[str, str]

config_id: str

config_name: str

created_at: datetime

custom_prediction_suffix: str | None

deployment_id: str

model_config: ClassVar[ConfigDict] = {'protected_namespaces': ()}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

model_name: str

property prediction_url

prediction_urls: ClassVar[PredictionURLs] = <gen_ai_hub.proxy.core.utils.PredictionURLs object>

Parameters:

model_name (str)
url (str)
fixed_suffix (Optional[str])

Return type:

str

url: str

class FoundationalModelScenario

Bases: BaseModel

Represents a foundational model scenario in the GenAI Hub.

classmethod adjust(data)

Adjust input data before model initialization.

Parameters:: data (Any) -- Input data to adjust.
Returns:: Adjusted data.
Return type:: Any

config_names: List[str] | str | None

model_config: ClassVar[ConfigDict] = {'protected_namespaces': ()}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

model_name_parameter: str

prediction_url_suffix: str | None

scenario_id: str

class GenAIHubProxyClient

Bases: BaseProxyClient

GenAIHubProxyClient is a proxy client for interacting with the GenAI Hub.

classmethod add_foundation_model_scenario(scenario_id, config_names=None, prediction_url_suffix=None, model_name_parameter='model_name')

Add a foundational model scenario to the client.

Parameters:

scenario_id (str) -- the scenario ID.
config_names (Optional[List[str]], optional) -- list of configuration names, defaults to None
prediction_url_suffix (Optional[str], optional) -- prediction URL suffix, defaults to None
model_name_parameter (str, optional) -- model name parameter, defaults to 'model_name'

classmethod for_profile(profile=None)

Create a GenAIHubProxyClient instance for the given profile.

Parameters:: profile (str, optional) -- Profile name, defaults to None
Returns:: GenAIHubProxyClient instance.
Return type:: GenAIHubProxyClient

classmethod init_client(data)

Initialize the client with the provided data.

Parameters:: data (Any) -- Input data for client initialization.
Returns:: Initialized data.
Return type:: Any

classmethod set_default_values(**kwargs): Set default values for the client.

get_additional_headers()

Get only the additional headers (instance-level and request-level).

Returns:: Additional headers.
Return type:: Dict[str, str]

get_ai_core_token()

Get the AI core token for authentication.

Returns:: AI core token.
Return type:: str

get_deployments()

Get the list of deployments.

Returns:: List of deployments.
Return type:: List[Deployment]

get_request_header()

Get the request headers for requests made by the client.

Returns:: Request headers.
Return type:: Dict[str, str]

model_post_init(context, /)

This function is meant to behave like a BaseModel method to initialise private attributes.

It takes context as an argument since that's what pydantic-core passes when calling it.

Args:: self: The BaseModel instance. context: The context.

Parameters:

self (BaseModel)
context (Any)

Return type:

None

select_deployment(raise_on_multiple=False, **search_key_value)

Parameters:: raise_on_multiple (bool)

set_headers_addition(headers)

Set additional headers for requests made by the client.

Parameters:: headers (Dict[str, str]) -- Headers to add.

update_deployments()

Update the list of deployments from the GenAI Hub.

Returns:: List of updated deployments.
Return type:: List[Deployment]

AI_CLIENT_TYPE_VAL: ClassVar[str] = 'GenAI Hub SDK (Python)'

ai_core_client: AICoreV2Client | None

auth_url: str | None

base_url: str | None

client_id: str | None

client_secret: str | None

default_values: ClassVar[Dict[str, Any]] = {}

property deployment_class: Type[Deployment]

property deployments: List[Deployment]

foundational_model_scenarios: ClassVar[List[FoundationalModelScenario]] = [FoundationalModelScenario(scenario_id='foundation-models', config_names=['*'], model_name_parameter='model_name', prediction_url_suffix=None)]

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True, 'extra': 'allow', 'protected_namespaces': ()}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

on_invalid_deployments: ClassVar[InvalidDeploymentBehavior] = 'warn'

property request_header: Dict[str, Any]

resource_group: str | None

class GenAIHubRestClient

Bases: object

REST client with automatic header injection.

This client wraps the AI Core rest_client and ensures that all requests include: - Instance-level headers (set via proxy_client.set_headers_addition) - Request-level headers (set via temporary_headers_addition context manager)

Parameters:: proxy_client -- The GenAIHubProxyClient instance to get the rest_client and headers from.

__init__(proxy_client)

Initialize the GenAIHubRestClient.

Parameters:: proxy_client (GenAIHubProxyClient) -- The GenAIHubProxyClient instance to get the rest_client and headers from.

delete(path, **kwargs)

Send a DELETE request with injected headers.

Parameters:

path (str) -- The API path.
kwargs -- Additional arguments to pass to the underlying rest_client.

Returns:

The response from the rest_client.

get(path, **kwargs)

Send a GET request with injected headers.

Parameters:

path (str) -- The API path.
kwargs -- Additional arguments to pass to the underlying rest_client.

Returns:

The response from the rest_client.

patch(path, **kwargs)

Send a PATCH request with injected headers.

Parameters:

path (str) -- The API path.
kwargs -- Additional arguments to pass to the underlying rest_client.

Returns:

The response from the rest_client.

post(path, **kwargs)

Send a POST request with injected headers.

Parameters:

path (str) -- The API path.
kwargs -- Additional arguments to pass to the underlying rest_client.

Returns:

The response from the rest_client.

class InvalidDeploymentBehavior

Bases: str, Enum

__new__(value)

ignore = 'ignore'

raise_error = 'raise_error'

warn = 'warn'

camel_to_snake(name)

Convert camelCase or PascalCase string to snake_case.

Parameters:: name (str) -- Input string in camelCase or PascalCase.
Returns:: String converted to snake_case.
Return type:: str

config_parameters(model_name_parameter, ai_core_client, deployment)

Get configuration parameters for a deployment.

Parameters:

model_name_parameter (str) -- the model name parameter.
ai_core_client (AICoreV2Client) -- the AI core client.
deployment (Deployment) -- the deployment.

Returns:

Dictionary with model name and additional parameters.

Return type:

Dict[str, Any]

temporary_headers_addition(headers)

Context manager to temporarily add headers to requests made by the GenAIHubProxyClient.

Parameters:: headers (Dict[str, str]) -- Headers to add temporarily.