aioli-sdk 1.2.0.dev5__tar.gz → 1.2.0.dev7__tar.gz

{aioli_sdk-1.2.0.dev5 → aioli_sdk-1.2.0.dev7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: aioli-sdk
-Version: 1.2.0.dev5
+Version: 1.2.0.dev7
 Summary: Aioli (AI OnLine Inference), a platform for deploying AI models at scale.
 Home-page: https://github.com/determined-ai/aioli
 Author: HPE AI Solutions

{aioli_sdk-1.2.0.dev5 → aioli_sdk-1.2.0.dev7}/aioli/__version__.py

@@ -1,2 +1,2 @@
 # © Copyright 2024 Hewlett Packard Enterprise Development LP
-__version__ = "1.2.0-dev5"
+__version__ = "1.2.0-dev7"

{aioli_sdk-1.2.0.dev5 → aioli_sdk-1.2.0.dev7}/aioli/cli/deployment.py

@@ -1,7 +1,7 @@
 # © Copyright 2023-2024 Hewlett Packard Enterprise Development LP
 import textwrap
 from argparse import Namespace
-from typing import Any, List
+from typing import Any, Dict, List
 
 import aiolirest
 from aioli import cli
@@ -184,6 +184,8 @@ def create(args: Namespace) -> None:
             canaryTrafficPercent=args.canary_traffic_percent,
             environment=construct_environment(args),
             arguments=construct_arguments(args),
+            nodeSelectors=construct_node_selectors(args),
+            priorityClassName=args.priority_class_name,
         )
         api_instance.deployments_post(r)
 
@@ -208,6 +210,10 @@ def update(args: Namespace) -> None:
             autoScaling=found.auto_scaling,
             canaryTrafficPercent=found.canary_traffic_percent,
             goalStatus=found.goal_status,
+            environment=found.environment,
+            arguments=found.arguments,
+            nodeSelectors=found.node_selectors,
+            priorityClassName=found.priority_class_name,
         )
 
         if request.auto_scaling is None:
@@ -258,6 +264,12 @@ def update(args: Namespace) -> None:
         if args.arg is not None:
             request.arguments = construct_arguments(args)
 
+        if args.node_selector is not None:
+            request.node_selectors = construct_node_selectors(args)
+
+        if args.priority_class_name is not None:
+            request.priority_class_name = args.priority_class_name
+
         headers = {"Content-Type": "application/json"}
         assert found.id is not None
         api_instance.deployments_id_put(found.id, request, _headers=headers)
@@ -318,6 +330,24 @@ def str2bool(v: str) -> bool:
         )
 
 
+def construct_node_selectors(args: Namespace) -> Dict[str, str]:
+    node_selectors: Dict[str, str] = {}
+    if args.node_selector is None:
+        return node_selectors
+
+    for entry in args.node_selector:
+        if entry is None:
+            continue
+        # split to label & value
+        the_split = entry.split("=", maxsplit=1)
+        label: str = the_split[0]
+        value: str = ""
+        if len(the_split) > 1:
+            value = the_split[1]
+        node_selectors[label] = value
+    return node_selectors
+
+
 common_deployment_args: ArgsDescription = [
     Arg(
         "--authentication-required",
@@ -345,17 +375,36 @@ common_deployment_args: ArgsDescription = [
     Arg(
         "-a",
         "--arg",
+        nargs="?",
         help="Argument to be added to the service command line. "
-        "If specifying an argument that starts with a '-', use the form --arg=<your-argument>",
+        "If specifying an argument that starts with a '-', use the form --arg=<your-argument>. "
+        "Specifying any --arg replaces prior args with the arguments on this invocation. "
+        "Use a single --arg with no value to clear all arguments.",
         action="append",
     ),
     Arg(
         "-e",
         "--env",
+        nargs="?",
         help="Specifies an environment variable & value as name=value, "
-        "to be passed to the launched container",
+        "to be passed to the launched container. "
+        "Specifying any --env replaces prior environment vars with those on this invocation. "
+        "Use a single --env with no value to clear all environment vars.",
         action="append",
     ),
+    Arg(
+        "--node-selector",
+        nargs="?",
+        help="Specifies a node selector label & value as label=value, "
+        "to be passed to the launched container. Example kubernetes.io/arch=amd64. "
+        "Specifying any --node-selector replaces prior selectors with those on this invocation. "
+        "Use a single --node-selector with no value to clear all selectors.",
+        action="append",
+    ),
+    Arg(
+        "--priority-class-name",
+        help="Priority Class Name to be used for prioritization of deployments",
+    ),
 ]
 
 main_cmd = Cmd(

{aioli_sdk-1.2.0.dev5 → aioli_sdk-1.2.0.dev7}/aioli/cli/model.py

@@ -343,15 +343,21 @@ common_model_args: ArgsDescription = [
     Arg(
         "-a",
         "--arg",
+        nargs="?",
         help="Argument to be added to the service command line. "
-        "If specifying an argument that starts with a '-', use the form --arg=<your-argument>",
+        "If specifying an argument that starts with a '-', use the form --arg=<your-argument>. "
+        "Specifying any --arg replaces prior args with the arguments on this invocation. "
+        "Use a single --arg with no value to clear all arguments.",
         action="append",
     ),
     Arg(
         "-e",
         "--env",
+        nargs="?",
         help="Specifies an environment variable & value as name=value, "
-        "to be passed to the launched container",
+        "to be passed to the launched container. "
+        "Specifying any --env replaces prior environment vars with those on this invocation. "
+        "Use a single --env with no value to clear all environment vars.",
         action="append",
     ),
     Arg("--gpu-type", help="GPU type required"),

{aioli_sdk-1.2.0.dev5 → aioli_sdk-1.2.0.dev7}/aioli/cli/test/test_cli.py

@@ -597,6 +597,7 @@ class TestCli:
         expected = (
             '[\n'    # noqa: Q000
             '  {\n'  # noqa: Q000
+            '    "arguments": [],\n'
             '    "autoScaling": {\n'
             '      "maxReplicas": 1,\n'
             '      "metric": "concurrency",\n'
@@ -652,6 +653,8 @@ class TestCli:
                 "--canary-traffic-percent 20 "
                 "-a='--debug' "
                 "-e MODS=SOME "
+                "--node-selector kubernetes.io/arch=amd64 "
+                "--priority-class-name '' "
                 "iris-tf-keras-deployment-2"
             )
             == 0
@@ -701,6 +704,7 @@ class TestCli:
         result_list = result.split("\n")
         args_found = False
         other_found = False
+        node_selector_found = False
         for line in result_list:
             if "- --updated" in line:
                 args_found = True
@@ -708,10 +712,94 @@ class TestCli:
                 mods_found = True
             if "OTHER: VALUE" in line:
                 other_found = True
+            if "kubernetes.io/arch: amd64" in line:
+                node_selector_found = True
 
         assert args_found
         assert mods_found
         assert other_found
+        assert node_selector_found
+
+    def test_deployment_update_node_selectors(self, setup_login: None) -> None:
+        # Update the deployments nodeSelectors check that changes
+        # were made. Also verify arguments and environment are preserved.
+        assert (
+            os.system(
+                "aioli deployment update "
+                "--node-selector x/y=a-b "
+                "--node-selector x/z=a-c "
+                "iris-tf-keras-deployment-2"
+            )
+            == 0
+        )
+
+        result = subprocess.check_output(
+            ["aioli", "deployment", "show", "iris-tf-keras-deployment-2"]
+        ).decode("utf-8")
+        result_list = result.split("\n")
+        args_found = False
+        other_found = False
+        prev_node_selector_not_found = True
+        node_selector_1_found = False
+        node_selector_2_found = False
+        for line in result_list:
+            if "- --updated" in line:
+                args_found = True
+            if "MODS: UPDATED" in line:
+                mods_found = True
+            if "OTHER: VALUE" in line:
+                other_found = True
+            if "kubernetes.io/arch: amd64" in line:
+                prev_node_selector_not_found = False
+            if "x/y: a-b" in line:
+                node_selector_1_found = True
+            if "x/z: a-c" in line:
+                node_selector_2_found = True
+        assert args_found
+        assert mods_found
+        assert other_found
+        assert node_selector_1_found
+        assert node_selector_2_found
+        assert prev_node_selector_not_found
+
+    def test_deployment_clear_env_args_node_selectors(self, setup_login: None) -> None:
+        # Update the deployments and clear env/args/node_selectors
+        assert (
+            os.system(
+                "aioli deployment update "
+                "iris-tf-keras-deployment-2 "
+                "--env "
+                "--arg "
+                "--node-selector"
+            )
+            == 0
+        )
+
+        result = subprocess.check_output(
+            ["aioli", "deployment", "show", "iris-tf-keras-deployment-2"]
+        ).decode("utf-8")
+        result_list = result.split("\n")
+        args_not_found = True
+        mods_not_found = True
+        other_not_found = True
+        node_selector_1_not_found = True
+        node_selector_2_not_found = True
+        for line in result_list:
+            if "- --updated" in line:
+                args_not_found = False
+            if "MODS: UPDATED" in line:
+                mods_not_found = False
+            if "OTHER: VALUE" in line:
+                other_not_found = False
+            if "x/y: a-b" in line:
+                node_selector_1_not_found = False
+            if "x/z: a-c" in line:
+                node_selector_2_not_found = False
+        assert args_not_found
+        assert mods_not_found
+        assert other_not_found
+        assert node_selector_1_not_found
+        assert node_selector_2_not_found
 
     def test_deployment_delete(self, setup_login: None) -> None:
         assert os.system("aioli deployment delete iris-tf-keras-deployment") == 0

{aioli_sdk-1.2.0.dev5 → aioli_sdk-1.2.0.dev7}/aioli/common/util.py

@@ -154,6 +154,8 @@ def construct_environment(args: Namespace) -> Dict[str, str]:
         return environment
 
     for entry in args.env:
+        if entry is None:
+            continue
         # split to name & value
         the_split = entry.split("=", maxsplit=1)
         name: str = the_split[0]
@@ -170,5 +172,7 @@ def construct_arguments(args: Namespace) -> List[str]:
         return arguments
 
     for entry in args.arg:
+        if entry is None:
+            continue
         arguments.append(entry.strip())
     return arguments

{aioli_sdk-1.2.0.dev5 → aioli_sdk-1.2.0.dev7}/aioli_sdk.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: aioli-sdk
-Version: 1.2.0.dev5
+Version: 1.2.0.dev7
 Summary: Aioli (AI OnLine Inference), a platform for deploying AI models at scale.
 Home-page: https://github.com/determined-ai/aioli
 Author: HPE AI Solutions

{aioli_sdk-1.2.0.dev5 → aioli_sdk-1.2.0.dev7}/aioli_sdk.egg-info/SOURCES.txt

@@ -62,6 +62,7 @@ aiolirest/models/autoscaling.py
 aiolirest/models/configuration_resources.py
 aiolirest/models/deployment.py
 aiolirest/models/deployment_model_version.py
+aiolirest/models/deployment_priority.py
 aiolirest/models/deployment_request.py
 aiolirest/models/deployment_state.py
 aiolirest/models/deployment_token.py

{aioli_sdk-1.2.0.dev5 → aioli_sdk-1.2.0.dev7}/aiolirest/__init__.py

@@ -7,7 +7,7 @@
 
     HPE MLIS is *Aioli* -- The AI On-line Inference Platform that enables easy deployment, tracking, and serving of your packaged models regardless of your preferred AI framework.
 
-    The version of the OpenAPI document: 1.2.0-dev5
+    The version of the OpenAPI document: 1.2.0-dev7
     Contact: community@determined-ai
     Generated by OpenAPI Generator (https://openapi-generator.tech)
 
@@ -45,6 +45,7 @@ from aiolirest.models.autoscaling import Autoscaling
 from aiolirest.models.configuration_resources import ConfigurationResources
 from aiolirest.models.deployment import Deployment
 from aiolirest.models.deployment_model_version import DeploymentModelVersion
+from aiolirest.models.deployment_priority import DeploymentPriority
 from aiolirest.models.deployment_request import DeploymentRequest
 from aiolirest.models.deployment_state import DeploymentState
 from aiolirest.models.deployment_token import DeploymentToken

{aioli_sdk-1.2.0.dev5 → aioli_sdk-1.2.0.dev7}/aiolirest/api/authentication_api.py

@@ -5,7 +5,7 @@
 
     HPE MLIS is *Aioli* -- The AI On-line Inference Platform that enables easy deployment, tracking, and serving of your packaged models regardless of your preferred AI framework.
 
-    The version of the OpenAPI document: 1.2.0-dev5
+    The version of the OpenAPI document: 1.2.0-dev7
     Contact: community@determined-ai
     Generated by OpenAPI Generator (https://openapi-generator.tech)
 

{aioli_sdk-1.2.0.dev5 → aioli_sdk-1.2.0.dev7}/aiolirest/api/deployments_api.py

@@ -5,7 +5,7 @@
 
     HPE MLIS is *Aioli* -- The AI On-line Inference Platform that enables easy deployment, tracking, and serving of your packaged models regardless of your preferred AI framework.
 
-    The version of the OpenAPI document: 1.2.0-dev5
+    The version of the OpenAPI document: 1.2.0-dev7
     Contact: community@determined-ai
     Generated by OpenAPI Generator (https://openapi-generator.tech)
 
@@ -31,6 +31,7 @@ from pydantic import StrictStr
 from typing import List, Optional
 
 from aiolirest.models.deployment import Deployment
+from aiolirest.models.deployment_priority import DeploymentPriority
 from aiolirest.models.deployment_request import DeploymentRequest
 from aiolirest.models.event_info import EventInfo
 from aiolirest.models.observability import Observability
@@ -73,7 +74,7 @@ class DeploymentsApi:
     ) -> List[Deployment]:
         """Get all deployments.
 
-        Return the properties of all configured deployments.
+        Return the properties of all configured deployments. Results are sorted by modifiedAt descending.
 
         :param last_event: Include the most recent event for each deployment when available.
         :type last_event: str
@@ -142,7 +143,7 @@ class DeploymentsApi:
     ) -> ApiResponse[List[Deployment]]:
         """Get all deployments.
 
-        Return the properties of all configured deployments.
+        Return the properties of all configured deployments. Results are sorted by modifiedAt descending.
 
         :param last_event: Include the most recent event for each deployment when available.
         :type last_event: str
@@ -211,7 +212,7 @@ class DeploymentsApi:
     ) -> RESTResponseType:
         """Get all deployments.
 
-        Return the properties of all configured deployments.
+        Return the properties of all configured deployments. Results are sorted by modifiedAt descending.
 
         :param last_event: Include the most recent event for each deployment when available.
         :type last_event: str
@@ -1729,7 +1730,7 @@ class DeploymentsApi:
         _headers: Optional[Dict[StrictStr, Any]] = None,
         _host_index: Annotated[StrictInt, Field(ge=0, le=0)] = 0,
     ) -> List[str]:
-        """Get all namespaces that can be selected for deployemnts.
+        """Get all namespaces that can be selected for deployments.
 
         List of namespace names.
 
@@ -1794,7 +1795,7 @@ class DeploymentsApi:
         _headers: Optional[Dict[StrictStr, Any]] = None,
         _host_index: Annotated[StrictInt, Field(ge=0, le=0)] = 0,
     ) -> ApiResponse[List[str]]:
-        """Get all namespaces that can be selected for deployemnts.
+        """Get all namespaces that can be selected for deployments.
 
         List of namespace names.
 
@@ -1859,7 +1860,7 @@ class DeploymentsApi:
         _headers: Optional[Dict[StrictStr, Any]] = None,
         _host_index: Annotated[StrictInt, Field(ge=0, le=0)] = 0,
     ) -> RESTResponseType:
-        """Get all namespaces that can be selected for deployemnts.
+        """Get all namespaces that can be selected for deployments.
 
         List of namespace names.
 
@@ -2242,3 +2243,253 @@ class DeploymentsApi:
         )
 
 
+
+
+    @validate_call
+    def deployments_priorities_get(
+        self,
+        _request_timeout: Union[
+            None,
+            Annotated[StrictFloat, Field(gt=0)],
+            Tuple[
+                Annotated[StrictFloat, Field(gt=0)],
+                Annotated[StrictFloat, Field(gt=0)]
+            ]
+        ] = None,
+        _request_auth: Optional[Dict[StrictStr, Any]] = None,
+        _content_type: Optional[StrictStr] = None,
+        _headers: Optional[Dict[StrictStr, Any]] = None,
+        _host_index: Annotated[StrictInt, Field(ge=0, le=0)] = 0,
+    ) -> List[DeploymentPriority]:
+        """Get all priority classes that can be selected for deployments.
+
+        List of priority classes.
+
+        :param _request_timeout: timeout setting for this request. If one
+                                 number provided, it will be total request
+                                 timeout. It can also be a pair (tuple) of
+                                 (connection, read) timeouts.
+        :type _request_timeout: int, tuple(int, int), optional
+        :param _request_auth: set to override the auth_settings for an a single
+                              request; this effectively ignores the
+                              authentication in the spec for a single request.
+        :type _request_auth: dict, optional
+        :param _content_type: force content-type for the request.
+        :type _content_type: str, Optional
+        :param _headers: set to override the headers for a single
+                         request; this effectively ignores the headers
+                         in the spec for a single request.
+        :type _headers: dict, optional
+        :param _host_index: set to override the host_index for a single
+                            request; this effectively ignores the host_index
+                            in the spec for a single request.
+        :type _host_index: int, optional
+        :return: Returns the result object.
+        """ # noqa: E501
+
+        _param = self._deployments_priorities_get_serialize(
+            _request_auth=_request_auth,
+            _content_type=_content_type,
+            _headers=_headers,
+            _host_index=_host_index
+        )
+
+        _response_types_map: Dict[str, Optional[str]] = {
+            '200': "List[DeploymentPriority]",
+            '500': "ErrorResponse"
+            
+        }
+        response_data = self.api_client.call_api(
+            *_param,
+            _request_timeout=_request_timeout
+        )
+        response_data.read()
+        return self.api_client.response_deserialize(
+            response_data=response_data,
+            response_types_map=_response_types_map,
+        ).data
+
+
+    @validate_call
+    def deployments_priorities_get_with_http_info(
+        self,
+        _request_timeout: Union[
+            None,
+            Annotated[StrictFloat, Field(gt=0)],
+            Tuple[
+                Annotated[StrictFloat, Field(gt=0)],
+                Annotated[StrictFloat, Field(gt=0)]
+            ]
+        ] = None,
+        _request_auth: Optional[Dict[StrictStr, Any]] = None,
+        _content_type: Optional[StrictStr] = None,
+        _headers: Optional[Dict[StrictStr, Any]] = None,
+        _host_index: Annotated[StrictInt, Field(ge=0, le=0)] = 0,
+    ) -> ApiResponse[List[DeploymentPriority]]:
+        """Get all priority classes that can be selected for deployments.
+
+        List of priority classes.
+
+        :param _request_timeout: timeout setting for this request. If one
+                                 number provided, it will be total request
+                                 timeout. It can also be a pair (tuple) of
+                                 (connection, read) timeouts.
+        :type _request_timeout: int, tuple(int, int), optional
+        :param _request_auth: set to override the auth_settings for an a single
+                              request; this effectively ignores the
+                              authentication in the spec for a single request.
+        :type _request_auth: dict, optional
+        :param _content_type: force content-type for the request.
+        :type _content_type: str, Optional
+        :param _headers: set to override the headers for a single
+                         request; this effectively ignores the headers
+                         in the spec for a single request.
+        :type _headers: dict, optional
+        :param _host_index: set to override the host_index for a single
+                            request; this effectively ignores the host_index
+                            in the spec for a single request.
+        :type _host_index: int, optional
+        :return: Returns the result object.
+        """ # noqa: E501
+
+        _param = self._deployments_priorities_get_serialize(
+            _request_auth=_request_auth,
+            _content_type=_content_type,
+            _headers=_headers,
+            _host_index=_host_index
+        )
+
+        _response_types_map: Dict[str, Optional[str]] = {
+            '200': "List[DeploymentPriority]",
+            '500': "ErrorResponse"
+            
+        }
+        response_data = self.api_client.call_api(
+            *_param,
+            _request_timeout=_request_timeout
+        )
+        response_data.read()
+        return self.api_client.response_deserialize(
+            response_data=response_data,
+            response_types_map=_response_types_map,
+        )
+
+
+    @validate_call
+    def deployments_priorities_get_without_preload_content(
+        self,
+        _request_timeout: Union[
+            None,
+            Annotated[StrictFloat, Field(gt=0)],
+            Tuple[
+                Annotated[StrictFloat, Field(gt=0)],
+                Annotated[StrictFloat, Field(gt=0)]
+            ]
+        ] = None,
+        _request_auth: Optional[Dict[StrictStr, Any]] = None,
+        _content_type: Optional[StrictStr] = None,
+        _headers: Optional[Dict[StrictStr, Any]] = None,
+        _host_index: Annotated[StrictInt, Field(ge=0, le=0)] = 0,
+    ) -> RESTResponseType:
+        """Get all priority classes that can be selected for deployments.
+
+        List of priority classes.
+
+        :param _request_timeout: timeout setting for this request. If one
+                                 number provided, it will be total request
+                                 timeout. It can also be a pair (tuple) of
+                                 (connection, read) timeouts.
+        :type _request_timeout: int, tuple(int, int), optional
+        :param _request_auth: set to override the auth_settings for an a single
+                              request; this effectively ignores the
+                              authentication in the spec for a single request.
+        :type _request_auth: dict, optional
+        :param _content_type: force content-type for the request.
+        :type _content_type: str, Optional
+        :param _headers: set to override the headers for a single
+                         request; this effectively ignores the headers
+                         in the spec for a single request.
+        :type _headers: dict, optional
+        :param _host_index: set to override the host_index for a single
+                            request; this effectively ignores the host_index
+                            in the spec for a single request.
+        :type _host_index: int, optional
+        :return: Returns the result object.
+        """ # noqa: E501
+
+        _param = self._deployments_priorities_get_serialize(
+            _request_auth=_request_auth,
+            _content_type=_content_type,
+            _headers=_headers,
+            _host_index=_host_index
+        )
+
+        _response_types_map: Dict[str, Optional[str]] = {
+            '200': "List[DeploymentPriority]",
+            '500': "ErrorResponse"
+            
+        }
+        response_data = self.api_client.call_api(
+            *_param,
+            _request_timeout=_request_timeout
+        )
+        return response_data.response
+
+
+    def _deployments_priorities_get_serialize(
+        self,
+        _request_auth,
+        _content_type,
+        _headers,
+        _host_index,
+    ) -> Tuple:
+
+        _host = None
+
+        _collection_formats: Dict[str, str] = {
+            
+        }
+
+        _path_params: Dict[str, str] = {}
+        _query_params: List[Tuple[str, str]] = []
+        _header_params: Dict[str, Optional[str]] = _headers or {}
+        _form_params: List[Tuple[str, str]] = []
+        _files: Dict[str, str] = {}
+        _body_params: Optional[bytes] = None
+
+        # process the path parameters
+        # process the query parameters
+        # process the header parameters
+        # process the form parameters
+        # process the body parameter
+
+
+        # set the HTTP header `Accept`
+        _header_params['Accept'] = self.api_client.select_header_accept(
+            [
+                '*/*'
+            ]
+        )
+
+
+        # authentication setting
+        _auth_settings: List[str] = [
+            'ApiKeyAuth'
+        ]
+
+        return self.api_client.param_serialize(
+            method='GET',
+            resource_path='/deployments/priorities',
+            path_params=_path_params,
+            query_params=_query_params,
+            header_params=_header_params,
+            body=_body_params,
+            post_params=_form_params,
+            files=_files,
+            auth_settings=_auth_settings,
+            collection_formats=_collection_formats,
+            _host=_host,
+            _request_auth=_request_auth
+        )
+
+

{aioli_sdk-1.2.0.dev5 → aioli_sdk-1.2.0.dev7}/aiolirest/api/information_api.py

@@ -5,7 +5,7 @@
 
     HPE MLIS is *Aioli* -- The AI On-line Inference Platform that enables easy deployment, tracking, and serving of your packaged models regardless of your preferred AI framework.
 
-    The version of the OpenAPI document: 1.2.0-dev5
+    The version of the OpenAPI document: 1.2.0-dev7
     Contact: community@determined-ai
     Generated by OpenAPI Generator (https://openapi-generator.tech)