-
Notifications
You must be signed in to change notification settings - Fork 706
/
data_types.py
92 lines (87 loc) · 4.33 KB
/
data_types.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
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
# Copyright 2020 Google LLC. All Rights Reserved.
#
# 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
#
# http://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.
"""Data types shared for orchestration."""
from typing import Any, Dict, List
import attr
from tfx import types
from tfx.orchestration import data_types_utils
from tfx.proto.orchestration import execution_invocation_pb2
from tfx.proto.orchestration import pipeline_pb2
# TODO(b/150979622): We should introduce an id that is not changed across
# retires of the same component run and pass it to executor operators for
# human-readability purpose.
# TODO(b/165359991): Restore 'auto_attribs=True' once we drop Python3.5 support.
@attr.s
class ExecutionInfo:
"""A struct to store information for an execution."""
# LINT.IfChange
# The Execution id that is registered in MLMD.
execution_id = attr.ib(type=int, default=None)
# The input map to feed to execution
input_dict = attr.ib(type=Dict[str, List[types.Artifact]], default=None)
# The output map to feed to execution
output_dict = attr.ib(type=Dict[str, List[types.Artifact]], default=None)
# The exec_properties to feed to execution
exec_properties = attr.ib(type=Dict[str, Any], default=None)
# The uri to execution result, note that the drivers or executors and
# Launchers may not run in the same process, so they should use this uri to
# "return" execution result to the launcher.
execution_output_uri = attr.ib(type=str, default=None)
# Stateful working dir will be deterministic given pipeline, node and run_id.
# The typical usecase is to restore long running executor's state after
# eviction. For examples, a Trainer can use this directory to store
# checkpoints.
stateful_working_dir = attr.ib(type=str, default=None)
# A tempory dir for executions and it is expected to be cleared up at the end
# of executions in both success and failure cases.
tmp_dir = attr.ib(type=str, default=None)
# The config of this Node.
pipeline_node = attr.ib(type=pipeline_pb2.PipelineNode, default=None)
# The config of the pipeline that this node is running in.
pipeline_info = attr.ib(type=pipeline_pb2.PipelineInfo, default=None)
# The id of the pipeline run that this execution is in.
pipeline_run_id = attr.ib(type=str, default=None)
# LINT.ThenChange(../../proto/orchestration/execution_invocation.proto)
def to_proto(self) -> execution_invocation_pb2.ExecutionInvocation:
return execution_invocation_pb2.ExecutionInvocation(
execution_id=self.execution_id,
input_dict=data_types_utils.build_artifact_struct_dict(self.input_dict),
output_dict=data_types_utils.build_artifact_struct_dict(
self.output_dict),
execution_properties=data_types_utils.build_metadata_value_dict(
self.exec_properties),
output_metadata_uri=self.execution_output_uri,
stateful_working_dir=self.stateful_working_dir,
tmp_dir=self.tmp_dir,
pipeline_node=self.pipeline_node,
pipeline_info=self.pipeline_info,
pipeline_run_id=self.pipeline_run_id)
@classmethod
def from_proto(
cls, execution_invocation: execution_invocation_pb2.ExecutionInvocation
) -> 'ExecutionInfo':
return cls(
execution_id=execution_invocation.execution_id,
input_dict=data_types_utils.build_artifact_dict(
execution_invocation.input_dict),
output_dict=data_types_utils.build_artifact_dict(
execution_invocation.output_dict),
exec_properties=data_types_utils.build_value_dict(
execution_invocation.execution_properties),
execution_output_uri=execution_invocation.output_metadata_uri,
stateful_working_dir=execution_invocation.stateful_working_dir,
tmp_dir=execution_invocation.tmp_dir,
pipeline_node=execution_invocation.pipeline_node,
pipeline_info=execution_invocation.pipeline_info,
pipeline_run_id=execution_invocation.pipeline_run_id)