/
kubeflow_metadata_adapter.py
62 lines (53 loc) · 2.37 KB
/
kubeflow_metadata_adapter.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
# Lint as: python2, python3
# Copyright 2019 Google LLC
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# https://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""A Metadata adapter class used to add Kubeflow-specific context."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import os
from typing import Any, Dict, Text
import absl
from tfx.orchestration import data_types
from tfx.orchestration import metadata
from ml_metadata.proto import metadata_store_pb2
_KFP_POD_NAME_ENV_KEY = 'KFP_POD_NAME'
_KFP_POD_NAME_PROPERTY_KEY = 'kfp_pod_name'
class KubeflowMetadataAdapter(metadata.Metadata):
"""A Metadata adapter class for pipelines run using KFP.
This is used to add properties to artifacts and executions, such as the Argo
pod IDs.
"""
def _is_eligible_previous_execution(
self, current_execution: metadata_store_pb2.Execution,
target_execution: metadata_store_pb2.Execution) -> bool:
current_execution.properties[_KFP_POD_NAME_PROPERTY_KEY].string_value = ''
target_execution.properties[_KFP_POD_NAME_PROPERTY_KEY].string_value = ''
return super(KubeflowMetadataAdapter,
self)._is_eligible_previous_execution(current_execution,
target_execution)
def _prepare_execution(
self,
state: Text,
exec_properties: Dict[Text, Any],
pipeline_info: data_types.PipelineInfo,
component_info: data_types.ComponentInfo,
) -> metadata_store_pb2.Execution:
if os.environ[_KFP_POD_NAME_ENV_KEY]:
kfp_pod_name = os.environ[_KFP_POD_NAME_ENV_KEY]
absl.logging.info('Adding KFP pod name %s to execution' % kfp_pod_name)
exec_properties[_KFP_POD_NAME_PROPERTY_KEY] = kfp_pod_name
return super(KubeflowMetadataAdapter,
self)._prepare_execution(state, exec_properties, pipeline_info,
component_info)