0
# ML Pipelines
1
2
Workflow orchestration, scheduling, and complex ML pipeline management with Kubeflow Pipelines integration.
3
4
## Capabilities
5
6
### Pipeline Jobs
7
8
Execute and manage ML workflows with comprehensive monitoring and scheduling.
9
10
```python { .api }
11
class PipelineJob:
12
@classmethod
13
def create(
14
cls,
15
display_name: str,
16
template_path: str,
17
job_id: Optional[str] = None,
18
pipeline_root: Optional[str] = None,
19
parameter_values: Optional[Dict[str, Any]] = None,
20
input_artifacts: Optional[Dict[str, str]] = None,
21
enable_caching: Optional[bool] = None,
22
encryption_spec_key_name: Optional[str] = None,
23
labels: Optional[Dict[str, str]] = None,
24
credentials: Optional[auth_credentials.Credentials] = None,
25
project: Optional[str] = None,
26
location: Optional[str] = None,
27
**kwargs
28
) -> 'PipelineJob': ...
29
30
def run(
31
self,
32
service_account: Optional[str] = None,
33
network: Optional[str] = None,
34
sync: bool = True,
35
create_request_timeout: Optional[float] = None,
36
**kwargs
37
) -> None: ...
38
39
@property
40
def state(self) -> PipelineState: ...
41
@property
42
def task_details(self) -> List[PipelineTaskDetail]: ...
43
```
44
45
### Pipeline Scheduling
46
47
Schedule pipeline executions with cron expressions and comprehensive scheduling options.
48
49
```python { .api }
50
class PipelineJobSchedule:
51
@classmethod
52
def create(
53
cls,
54
pipeline_job: PipelineJob,
55
display_name: str,
56
cron: str,
57
max_concurrent_run_count: int = 1,
58
max_run_count: Optional[int] = None,
59
service_account: Optional[str] = None,
60
network: Optional[str] = None,
61
labels: Optional[Dict[str, str]] = None,
62
**kwargs
63
) -> 'PipelineJobSchedule': ...
64
65
def pause(self) -> None: ...
66
def resume(self) -> None: ...
67
def update(self, **kwargs) -> 'PipelineJobSchedule': ...
68
```