-
Notifications
You must be signed in to change notification settings - Fork 1.2k
fix: update pyproject.toml dependencies for vector processing #3555
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
fix: update pyproject.toml dependencies for vector processing #3555
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm confused, the starter image has:
- provider_id: rag-runtime
provider_type: inline::rag-runtime
which pulls both pypdf
and chardet
.
However, I do agree that this is a miss. Especially if someone is not using inline::rag-runtime
.
Btw using started I don't repro your issue.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
can these be put in llama_stack/providers/registry/vector_io.py ?
for all providers? EDIT: at least the ones that speak openai vector store |
yes, make them deps for providers that use them |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
for Matt's comment
Thanks all for your comments. Yes, I encountered this when I tried to use openai compatible endpoint for file upload flow without rag runtime. |
Hi @mattf , @ashwinb , @leseb, I will go ahead and add chardet and pypdf to their pip_packages for all the following vector IO providers that have
|
9a2bf5d
to
ddfd2ef
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
please remove the changes in uv.lock and we can move fw with this
e166e47
to
9f036e7
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
please remove uv.lock changes
9f036e7
to
be35185
Compare
…egistries Moves chardet and pypdf from main dependencies to provider-specific pip_packages for all 12 vector IO providers that support file processing (Api.files). Updated providers: - inline::meta-reference, inline::faiss, inline::sqlite-vec, inline::sqlite_vec - remote::chromadb, inline::chromadb, remote::pgvector, remote::weaviate - inline::qdrant, remote::qdrant, remote::milvus, inline::milvus
05c36e1
to
d4466e3
Compare
Consolidate the common vector IO dependencies (chardet, pypdf) into a single DEFAULT_VECTOR_IO_DEPS variable to reduce duplication across all vector IO provider registries and the RAG tool runtime.
d4466e3
to
6c422ca
Compare
Hi @leseb , I took care of uv.lock comment you pointed out. Thanks. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
victory!
What does this PR do?
Updates pyproject.toml dependencies to fix vector processing compatibility issues.
closes: #3495
Test Plan
Tested llama stack server with faiss vector database:
- Created vector store with faiss backend
- Added PDF to vector store
- Performed semantic search queries