Added new sql vector functions #361
Conversation
Summary of ChangesHello @lvca, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed! This pull request significantly expands the SQL query language by introducing a comprehensive set of new vector functions. These additions are designed to empower users with advanced capabilities for similarity search, embeddings processing, and vector analytics directly within SQL queries. The functions cover a wide range of operations, from basic arithmetic and normalization to sophisticated techniques like quantization, sparse vector handling, and hybrid search reranking, providing a robust toolkit for modern data applications. Highlights
🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console. Using Gemini Code AssistThe full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips. Invoking Gemini You can request assistance from Gemini at any point by creating a comment using either
Customization To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a Limitations & Feedback Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here. You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension. Footnotes
|
🧪 CI InsightsHere's what we observed from your CI run for be4bc98. 🟢 All jobs passed!But CI Insights is watching 👀 |
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request introduces extensive documentation for a new set of SQL vector functions. The documentation is well-structured and comprehensive, covering a wide range of operations from basic arithmetic to advanced features like quantization and sparse vectors. My review focuses on ensuring the accuracy and clarity of the provided examples and descriptions. I've identified a few minor issues, such as an incorrect range for cosine similarity, a potential miscalculation in an example, and areas where more clarification would be beneficial for the user. Overall, this is a great addition to the documentation.
| ====== vectorInnerProduct() | ||
|
|
||
| Calculates the inner product of two vectors. | ||
| Can be negative. Primarily used with normalized vectors where it equals cosine similarity * magnitude product. |
There was a problem hiding this comment.
The function vectorInnerProduct appears to have the same functionality as vectorDotProduct. The documentation for vectorDotProduct even mentions "(inner product)" in its description. This duplication can be confusing for users. If they are aliases, it would be helpful to state that explicitly. If there is a subtle difference in implementation or use case (e.g., performance characteristics, as hinted by the SIMD optimization note for vectorDotProduct), it should be clearly documented to guide users on when to use which function.
| SELECT multiVectorScore([0.9, 0.8, 0.85], 'WEIGHTED', [0.5, 0.3, 0.2]) as weighted_score | ||
| ---- | ||
|
|
||
| Result: `0.87` |
There was a problem hiding this comment.
The result for the WEIGHTED example of multiVectorScore seems to be incorrect. The calculation (0.9 * 0.5) + (0.8 * 0.3) + (0.85 * 0.2) equals 0.45 + 0.24 + 0.17 = 0.86. The documentation states the result is 0.87. Please correct the example result.
Result: `0.86`
Update all 26 vector function references in SQL documentation to match the new camelCase naming convention. Changes: - Updated all UPPER_CASE_WITH_UNDERSCORE function names to camelCase - Updated all function headers (====== sections) - Updated all function signatures and examples - 33 vector function documentation sections updated Examples: - VECTOR_AVG() → vectorAvg() - VECTOR_SUM() → vectorSum() - SPARSE_VECTOR_CREATE() → sparseVectorCreate() - VECTOR_QUANTIZE_INT8() → vectorQuantizeInt8() - VECTOR_RRF_SCORE() → vectorRRFScore() Documentation remains accurate with function behavior unchanged, only naming convention updated to match ArcadeDB standards.
1 participant
No description provided.