hanzo-ml/trainer
1
0
Fork 0
mirror of https://github.com/hanzo-ml/trainer.git synced 2026-09-24 00:31:34 +00:00
Code Issues Packages Projects Releases Wiki Activity

chore(operator): Bump JobSet to v0.9.0 version (#2821)

Browse Source 
* chore(operator): Bump JobSet to v0.9.0 version

Signed-off-by: Andrey Velichkevich <[email protected]>

* Fix Go tests

Signed-off-by: Andrey Velichkevich <[email protected]>

---------

Signed-off-by: Andrey Velichkevich <[email protected]>
This commit is contained in:
Andrey Velichkevich
2025-09-09 06:55:36 +00:00
committed by GitHub
parent 6bbccf65c3
commit ca158302e7
14 changed files with 148 additions and 83 deletions
Download Patch File Download Diff File Expand all files Collapse all files
api/python_api/kubeflow_trainer_api/models/jobset_v1alpha2_job_set_status.py
+3 -3
View File
@@ -30,9 +30,9 @@ class JobsetV1alpha2JobSetStatus(BaseModel):
""" # noqa: E501
conditions: Optional[List[IoK8sApimachineryPkgApisMetaV1Condition]] = None
replicated_jobs_status: Optional[List[JobsetV1alpha2ReplicatedJobStatus]] = Field(default=None, description="ReplicatedJobsStatus track the number of JobsReady for each replicatedJob.", alias="replicatedJobsStatus")
restarts: Optional[StrictInt] = Field(default=None, description="Restarts tracks the number of times the JobSet has restarted (i.e. recreated in case of RecreateAll policy).")
restarts: Optional[StrictInt] = Field(default=0, description="Restarts tracks the number of times the JobSet has restarted (i.e. recreated in case of RecreateAll policy).")
restarts_count_towards_max: Optional[StrictInt] = Field(default=None, description="RestartsCountTowardsMax tracks the number of times the JobSet has restarted that counts towards the maximum allowed number of restarts.", alias="restartsCountTowardsMax")
terminal_state: Optional[StrictStr] = Field(default=None, description="TerminalState the state of the JobSet when it finishes execution. It can be either Complete or Failed. Otherwise, it is empty by default.", alias="terminalState")
terminal_state: Optional[StrictStr] = Field(default=None, description="TerminalState the state of the JobSet when it finishes execution. It can be either Completed or Failed. Otherwise, it is empty by default.", alias="terminalState")
__properties: ClassVar[List[str]] = ["conditions", "replicatedJobsStatus", "restarts", "restartsCountTowardsMax", "terminalState"]
model_config = ConfigDict(
@@ -102,7 +102,7 @@ class JobsetV1alpha2JobSetStatus(BaseModel):
_obj = cls.model_validate({
"conditions": [IoK8sApimachineryPkgApisMetaV1Condition.from_dict(_item) for _item in obj["conditions"]] if obj.get("conditions") is not None else None,
"replicatedJobsStatus": [JobsetV1alpha2ReplicatedJobStatus.from_dict(_item) for _item in obj["replicatedJobsStatus"]] if obj.get("replicatedJobsStatus") is not None else None,
"restarts": obj.get("restarts"),
"restarts": obj.get("restarts") if obj.get("restarts") is not None else 0,
"restartsCountTowardsMax": obj.get("restartsCountTowardsMax"),
"terminalState": obj.get("terminalState")
})
api/python_api/kubeflow_trainer_api/models/jobset_v1alpha2_replicated_job.py
+3 -1
View File
@@ -29,10 +29,11 @@ class JobsetV1alpha2ReplicatedJob(BaseModel):
JobsetV1alpha2ReplicatedJob
""" # noqa: E501
depends_on: Optional[List[JobsetV1alpha2DependsOn]] = Field(default=None, description="DependsOn is an optional list that specifies the preceding ReplicatedJobs upon which the current ReplicatedJob depends. If specified, the ReplicatedJob will be created only after the referenced ReplicatedJobs reach their desired state. The Order of ReplicatedJobs is defined by their enumeration in the slice. Note, that the first ReplicatedJob in the slice cannot use the DependsOn API. Currently, only a single item is supported in the DependsOn list. If JobSet is suspended the all active ReplicatedJobs will be suspended. When JobSet is resumed the Job sequence starts again. This API is mutually exclusive with the StartupPolicy API.", alias="dependsOn")
group_name: Optional[StrictStr] = Field(default=None, description="GroupName defines the name of the group this ReplicatedJob belongs to. Defaults to \"default\"", alias="groupName")
name: StrictStr = Field(description="Name is the name of the entry and will be used as a suffix for the Job name.")
replicas: Optional[StrictInt] = Field(default=None, description="Replicas is the number of jobs that will be created from this ReplicatedJob's template. Jobs names will be in the format: <jobSet.name>-<spec.replicatedJob.name>-<job-index>")
template: IoK8sApiBatchV1JobTemplateSpec = Field(description="Template defines the template of the Job that will be created.")
__properties: ClassVar[List[str]] = ["dependsOn", "name", "replicas", "template"]
__properties: ClassVar[List[str]] = ["dependsOn", "groupName", "name", "replicas", "template"]
model_config = ConfigDict(
populate_by_name=True,
@@ -96,6 +97,7 @@ class JobsetV1alpha2ReplicatedJob(BaseModel):
_obj = cls.model_validate({
"dependsOn": [JobsetV1alpha2DependsOn.from_dict(_item) for _item in obj["dependsOn"]] if obj.get("dependsOn") is not None else None,
"groupName": obj.get("groupName"),
"name": obj.get("name") if obj.get("name") is not None else '',
"replicas": obj.get("replicas"),
"template": IoK8sApiBatchV1JobTemplateSpec.from_dict(obj["template"]) if obj.get("template") is not None else None