Split text entries by max tokens supported by ML models #105
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
- Issue ML Models truncate entries exceeding some max token limit. This lowers the quality of search results - Fix Split entries by max tokens before indexing. This should improve searching for content in longer entries. - Miscellaneous - Test method to split entries by max tokens
- Test usage the entry splitting by max tokens in text search
- Fix logger message when converting org node to entries - Remove unused import from conftest
- Required because entries are now split by the max_word count supported by the ML models - This would now result in potentially duplicate hits, entries being returned to user - Do deduplication after ranking to get the top ranked deduplicated results
…ug logs - Remove property drawer from test entry for max_words splitting test - Property drawer is not required for the test - Keep minimal test case to reduce chance for confusion
debanjum
force-pushed
the
chunk-entries-by-max-token-size
branch
from
d5b340c to
24676f9
Compare
debanjum
added
upgrade
debanjum
changed the title
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Outdated suggestions cannot be applied.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.
Background
There is a limit to the maximum input tokens (words) that an ML model can encode into an embedding vector.
For the models used for text search in khoj, a max token size of 256 words is appropriate 1,2
Issue
Until now entries exceeding max token size would silently get truncated during embedding generation.
So the truncated portion of the entries would be ignored when matching queries with entries
This would degrade the quality of the results
Fix
Results