Skip to content
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

Make it possible to use chromadb in client-server mode #17

Closed
tazarov opened this issue Jul 19, 2023 · 0 comments
Closed

Make it possible to use chromadb in client-server mode #17

tazarov opened this issue Jul 19, 2023 · 0 comments
Labels
enhancement New feature or request
Milestone

Comments

@tazarov
Copy link
Contributor

tazarov commented Jul 19, 2023

About

Add http client support:

import chromadb
client = chromadb.HttpClient(host="localhost", port="8000")
@tazarov tazarov added the enhancement New feature or request label Jul 19, 2023
@tazarov tazarov added this to the POC milestone Jul 19, 2023
tazarov added a commit that referenced this issue Jul 19, 2023
- Added support for any chroma client (http, persistent etc)
- It is now possible to pass embedding function to use with the collection in chroma (Note: this may not be ideal if somehow different embeddings are used in different runs)
- Function Indexer now supports default LLM interface used for function wrappers
- All indexed functions are now loaded at Function Indexer instantiation
- Function searching now returns named tuple SearchResult (['name', 'wrapper', 'function', 'distance'])
- Improved the way a function is called from the wrapper to also accept *args in addition to **kwargs
- Each wrapped function now gets an identifier (e.g. `tests.test_findex.TestClass.test_class_method`) to be able to find reference to the function at Function Indexer startup. We also add a hash wich is the SHA-1 of the identifier
- Nested functions are not supported in the wrapper.
- It is not possible to rehydrate partial functions that have been indexed. Instead at app startup such function should be re-added to the index (not to worry the indexer will check the hashes and will not attempt to add an existing function)
- Updated docs for function indexing.

BREAKING-CHANGE: The `FunctionIndexer` init and other method signatures have changed and are not backward compatible.

Refs: #17
tazarov added a commit that referenced this issue Jul 19, 2023
* feat: Function indexer improvements

- Added support for any chroma client (http, persistent etc)
- It is now possible to pass embedding function to use with the collection in chroma (Note: this may not be ideal if somehow different embeddings are used in different runs)
- Function Indexer now supports default LLM interface used for function wrappers
- All indexed functions are now loaded at Function Indexer instantiation
- Function searching now returns named tuple SearchResult (['name', 'wrapper', 'function', 'distance'])
- Improved the way a function is called from the wrapper to also accept *args in addition to **kwargs
- Each wrapped function now gets an identifier (e.g. `tests.test_findex.TestClass.test_class_method`) to be able to find reference to the function at Function Indexer startup. We also add a hash wich is the SHA-1 of the identifier
- Nested functions are not supported in the wrapper.
- It is not possible to rehydrate partial functions that have been indexed. Instead at app startup such function should be re-added to the index (not to worry the indexer will check the hashes and will not attempt to add an existing function)
- Updated docs for function indexing.

BREAKING-CHANGE: The `FunctionIndexer` init and other method signatures have changed and are not backward compatible.

Refs: #17

---------

Signed-off-by: Trayan Azarov <tazarov@users.noreply.github.com>
@tazarov tazarov closed this as completed Jul 19, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

1 participant