Update ChromaDB tutorial materials for ChromaDB 1.5 and openai 3.x - #831
Open
realpython-bot wants to merge 3 commits into
Open
realpython-bot wants to merge 3 commits into
realpython-bot wants to merge 3 commits into
Conversation
- Pin all dependencies to the tested versions (Python 3.12+)
- Use configuration={"hnsw": ...} instead of metadata hnsw:space
- Fix off-by-one that dropped the last document of every batch and
read the batch size from client.get_max_batch_size()
- Use itertools.batched() instead of more-itertools
- Load the collection from its persisted embedding function config
- Polars: schema_overrides= and maintain_order=True
- Move the OpenAI code to the OpenAI() client and Responses API,
with the API key loaded from .env via python-dotenv
- Use the spaCy large model, since the 3.8 medium model prunes vectors
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- chroma_utils.py: import batched from itertools, not more_itertools. more-itertools is not in requirements.txt, so the script raised ImportError for anyone following the README. The tutorial uses the stdlib version, which the PR description already claimed. - car_data_etl.py: rename Vehicle_Model to Vehicle_Make. Index 1 of a title like "2017 Volvo XC90 ..." is the make, and the tutorial documents the metadata key as Vehicle_Make, so the collection built from this script had a different key than the article shows. - README.md: use the house (.venv) prompts and pin the spaCy model to en_core_web_lg-3.8.0, matching the tutorial's install commands. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
3 tasks
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
Supporting code for the full update of Embeddings and Vector Databases With ChromaDB (CMS update draft: Post 2390, Trello card https://trello.com/c/e5bIvapv).
Changes
chroma_utils.py: useconfiguration={"hnsw": {"space": ...}}instead of the oldmetadata={"hnsw:space": ...}, read the batch size fromclient.get_max_batch_size(), and fix the off-by-one (end_idx = batch[-1] + 1) that dropped the last document of every batch. Useitertools.batched()instead of more-itertools.car_data_etl.py:schema_overrides=(thedtypes=argument is deprecated in Polars 1.x) andmaintain_order=Trueso the printed example review is reproducible.create_car_review_collection.py: load the collection through its persisted embedding-function configuration and printcollection.count().llm_car_review_context.py: replace the removedopenai.ChatCompletioninterface with theOpenAI()client and the Responses API; loadOPENAI_API_KEYfrom.envwith python-dotenv (dropsconfig.json).word_vectors.py: useen_core_web_lg, because the 3.8 medium model prunes its vectors so that dog and cat share one vector..envinstructions.Testing
All scripts ran end to end on Python 3.14 with the pinned versions, including the full 5,870-review collection build and live OpenAI calls. The outputs match the updated tutorial draft.
🤖 Generated with Claude Code