Batch processing in training of NN ensemble - base project suggest calls#676
Draft
juhoinkinen wants to merge 4 commits into
Draft
Batch processing in training of NN ensemble - base project suggest calls#676juhoinkinen wants to merge 4 commits into
juhoinkinen wants to merge 4 commits into
Conversation
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #676 +/- ##
=======================================
Coverage 99.57% 99.57%
=======================================
Files 87 87
Lines 6157 6164 +7
=======================================
+ Hits 6131 6138 +7
Misses 26 26 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
|
Kudos, SonarCloud Quality Gate passed!
|
This was referenced Apr 14, 2023
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR experiments with implementing batched suggest calls for the base projects in NN ensemble backend.
Unfortunately there is no notable performance gain in real use, at least with MLLM, fastText, and Omikuji base projects (as in YSO projects of Finto AI), but actually a performance regression. Performance gain is seen when using only Omikuji as the base project, which is the only one of the backends in Finto AI YSO base models having the batch suggest method implemented.
Below results are from for runs at kj-kk using 16 jobs training on
corpora/fulltext-train/fi/*/.MLLM, fastText, and Omikuji base projects
1000 docs, 1 epoch
2000 docs, 10 epochs
Omikuji base project only
2000 docs, 1 epoch