Skip to content

Commit 92f4b40

Browse files
authored
Fix late initialization field (#165)
Issue #, if available: When a distributed/parallelized pytorch workload is run, AlgorithmSpecification.MetricDefinitions does not late initialize which makes the controller re queue for a long time. Description of changes: Edited generator.yaml to remove late_initialize and to ignore comparisons(ACK trainingjob does not support update operations so this wont effect the user). By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
1 parent 1c67e4e commit 92f4b40

File tree

5 files changed

+7
-20
lines changed

5 files changed

+7
-20
lines changed
Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,13 +1,13 @@
11
ack_generate_info:
2-
build_date: "2022-08-12T20:32:53Z"
2+
build_date: "2022-08-17T21:49:11Z"
33
build_hash: fe61d04673fd4d9848d5f726b01e0689a16d3733
44
go_version: go1.17.1
55
version: v0.19.3-1-gfe61d04
66
api_directory_checksum: 8b3c128d2037d5227679cccb57cd4d78af6aed1b
77
api_version: v1alpha1
88
aws_sdk_go_version: v1.42.28
99
generator_config_info:
10-
file_checksum: 92038dd37f8302115ea2733a22973f551e60c78e
10+
file_checksum: b97e2519e420215d25350e3df27f956c482e828e
1111
original_file_name: generator.yaml
1212
last_modification:
1313
reason: API generation

apis/v1alpha1/generator.yaml

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -171,8 +171,8 @@ resources:
171171
operation: DescribeTrainingJob
172172
path: FailureReason
173173
AlgorithmSpecification.MetricDefinitions:
174-
late_initialize:
175-
min_backoff_seconds: 5
174+
compare:
175+
is_ignored: true
176176
EnableInterContainerTrafficEncryption:
177177
late_initialize:
178178
min_backoff_seconds: 5

generator.yaml

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -171,8 +171,8 @@ resources:
171171
operation: DescribeTrainingJob
172172
path: FailureReason
173173
AlgorithmSpecification.MetricDefinitions:
174-
late_initialize:
175-
min_backoff_seconds: 5
174+
compare:
175+
is_ignored: true
176176
EnableInterContainerTrafficEncryption:
177177
late_initialize:
178178
min_backoff_seconds: 5

pkg/resource/training_job/delta.go

Lines changed: 0 additions & 3 deletions
Some generated files are not rendered by default. Learn more about customizing how changed files appear on GitHub.

pkg/resource/training_job/manager.go

Lines changed: 1 addition & 11 deletions
Some generated files are not rendered by default. Learn more about customizing how changed files appear on GitHub.

0 commit comments

Comments
 (0)