Refactor and make HealthLLM.jl more modular and easy to use - #21
Refactor and make HealthLLM.jl more modular and easy to use#21ParamThakkar123 wants to merge 20 commits into
Conversation
TheCedarPrince
left a comment
There was a problem hiding this comment.
Let's keep the conversation going here.
- Resolved merge conflicts in .gitignore, Project.toml, docs/src/index.md, src/HealthLLM.jl, src/database.jl, src/query.jl, src/utils.jl, test/runtests.jl - Removed src/embedding.jl and src/pgvector.jl (deleted by other branch; functionality inlined into utils.jl and database.jl respectively) - Added missing imports to HealthLLM module - Added docstrings with arguments, returns, and examples to public functions - Updated docs example to show full pipeline
…test/Project.toml
- docs/src/index.md: Full pipeline guide covering file collection, model registration, embedding building/validation/storage, RAG querying, and database query usage - src/utils.jl: Add docstrings to build_index_rag, get_schema, HuggingFaceLoadResult, load_huggingface_model, register_models - test/: Fix UtilsTest, DatabaseTest, QueryTest with proper module imports; remove stale PgvectorTest/EmbeddingTest; add pipeline tests; remove DrWatson dependency; fix test/Project.toml deps - Project.toml: Add DrWatson compat entry for Aqua compliance
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #21 +/- ##
===========================================
+ Coverage 27.27% 60.78% +33.51%
===========================================
Files 4 4
Lines 88 102 +14
===========================================
+ Hits 24 62 +38
+ Misses 64 40 -24 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
TheCedarPrince
left a comment
There was a problem hiding this comment.
Quick comment @ParamThakkar123 :
Could you give me an executive summary in your own words of what all you did here as well as how I should use this and test it out?
|
Sure @TheCedarPrince
|
|
And I have also added a script for you to test the pipeline, you will need ollama and pgvector setup to run the script. the script is present in the root folder as |
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

No description provided.