Fix inference logic and standardize config index mapping #1284
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This commit addresses several issues and improvements in the inference logic and config index mapping: 1. Fixes in RealTimeInferencer: * Previously, we checked if the last update time of the model state was within the current interval and skipped inference if it was. However, this led to excessive skipping of inference because the last update time was updated when retrieving the model state from the cache. * Introduced lastSeenExecutionEndTime in the model state, which specifically tracks the last time a sample was processed during inference (not training). This ensures more accurate control over when inference should be skipped. 2. Consistent Naming in Config Index Mapping: * To maintain consistency across the codebase, changed defaultFill to default_fill in the Config index mapping, following the underscore naming convention used elsewhere. 3. Additional Null Checks: * Added more null checks for the defaultFill field in the Config constructor to improve robustness. Testing: * Added a smoke test to allow the job scheduler to trigger anomaly detection inferencing, successfully reproducing and verifying the fix for item #1.* added unit tests for item #3. Signed-off-by: Kaituo Li <kaituo@amazon.com>
kaituo
requested review from
jmazanec15,
jngz-es,
saratvemulapalli,
ohltyler,
vamshin,
VijayanB,
ylwu-amzn,
amitgalitz,
jackiehanyang,
sean-zheng-amazon,
dbwiddis,
owaiskazi19 and
joshpalis
as code owners
opensearch-trigger-bot
bot
added
infra
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #1284 +/- ##
============================================
+ Coverage 71.83% 77.56% +5.73%
- Complexity 4898 5428 +530
============================================
Files 518 532 +14
Lines 22879 23260 +381
Branches 2245 2302 +57
============================================
+ Hits 16434 18041 +1607
+ Misses 5410 4169 -1241
- Partials 1035 1050 +15
Flags with carried forward coverage won't be shown. Click here to find out more.
|
|
bwc test failed due to search failure. Looks like core issue. Updated opensearch-project/OpenSearch#15234 |
|
WhiteSource security check failed as usual. |
amitgalitz
reviewed
Aug 23, 2024
| import com.google.gson.JsonObject; | ||
|
|
||
| /** | ||
| * Test that is meant to run with job scheduler to test if we have at least consecutive results generated. |
amitgalitz
approved these changes
Aug 23, 2024
opensearch-trigger-bot bot
pushed a commit
that referenced
this pull request
Aug 23, 2024
This commit addresses several issues and improvements in the inference logic and config index mapping: 1. Fixes in RealTimeInferencer: * Previously, we checked if the last update time of the model state was within the current interval and skipped inference if it was. However, this led to excessive skipping of inference because the last update time was updated when retrieving the model state from the cache. * Introduced lastSeenExecutionEndTime in the model state, which specifically tracks the last time a sample was processed during inference (not training). This ensures more accurate control over when inference should be skipped. 2. Consistent Naming in Config Index Mapping: * To maintain consistency across the codebase, changed defaultFill to default_fill in the Config index mapping, following the underscore naming convention used elsewhere. 3. Additional Null Checks: * Added more null checks for the defaultFill field in the Config constructor to improve robustness. Testing: * Added a smoke test to allow the job scheduler to trigger anomaly detection inferencing, successfully reproducing and verifying the fix for item #1.* added unit tests for item #3. Signed-off-by: Kaituo Li <kaituo@amazon.com> (cherry picked from commit 2922bbd) Signed-off-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
kaituo
added
bug
kaituo
pushed a commit
that referenced
this pull request
Aug 23, 2024
This commit addresses several issues and improvements in the inference logic and config index mapping: 1. Fixes in RealTimeInferencer: * Previously, we checked if the last update time of the model state was within the current interval and skipped inference if it was. However, this led to excessive skipping of inference because the last update time was updated when retrieving the model state from the cache. * Introduced lastSeenExecutionEndTime in the model state, which specifically tracks the last time a sample was processed during inference (not training). This ensures more accurate control over when inference should be skipped. 2. Consistent Naming in Config Index Mapping: * To maintain consistency across the codebase, changed defaultFill to default_fill in the Config index mapping, following the underscore naming convention used elsewhere. 3. Additional Null Checks: * Added more null checks for the defaultFill field in the Config constructor to improve robustness. Testing: * Added a smoke test to allow the job scheduler to trigger anomaly detection inferencing, successfully reproducing and verifying the fix for item #1.* added unit tests for item #3. Signed-off-by: Kaituo Li <kaituo@amazon.com> (cherry picked from commit 2922bbd) Signed-off-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
kaituo
pushed a commit
that referenced
this pull request
Aug 26, 2024
This commit addresses several issues and improvements in the inference logic and config index mapping: 1. Fixes in RealTimeInferencer: * Previously, we checked if the last update time of the model state was within the current interval and skipped inference if it was. However, this led to excessive skipping of inference because the last update time was updated when retrieving the model state from the cache. * Introduced lastSeenExecutionEndTime in the model state, which specifically tracks the last time a sample was processed during inference (not training). This ensures more accurate control over when inference should be skipped. 2. Consistent Naming in Config Index Mapping: * To maintain consistency across the codebase, changed defaultFill to default_fill in the Config index mapping, following the underscore naming convention used elsewhere. 3. Additional Null Checks: * Added more null checks for the defaultFill field in the Config constructor to improve robustness. Testing: * Added a smoke test to allow the job scheduler to trigger anomaly detection inferencing, successfully reproducing and verifying the fix for item #1.* added unit tests for item #3. (cherry picked from commit 2922bbd) Signed-off-by: Kaituo Li <kaituo@amazon.com> Signed-off-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com> Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
This commit addresses several issues and improvements in the inference logic and config index mapping:
Testing:
Check List
--signoff.By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.