Commit Graph

124 Commits

Author SHA1 Message Date
Andy Lee
9a5c197acd fix: auto-detect normalized embeddings and use cosine distance
- Add automatic detection for normalized embedding models (OpenAI, Voyage AI, Cohere)
- Automatically set distance_metric='cosine' for normalized embeddings
- Add warnings when using non-optimal distance metrics
- Implement manual L2 normalization in HNSW backend (custom Faiss build lacks normalize_L2)
- Fix DiskANN zmq_port compatibility with lazy loading strategy
- Add documentation for normalized embeddings feature

This fixes the low accuracy issue when using OpenAI text-embedding-3-small model with default MIPS metric.
2025-07-27 20:21:05 -07:00
yichuan520030910320
af1790395a fix ruff errors and formatting 2025-07-27 02:22:54 -07:00
GitHub Actions
5d09586853 chore: release v0.1.14 2025-07-27 08:50:56 +00:00
Andy Lee
a7cba078dd chore: consolidate essential fixes and add pre-commit hooks
- Add pre-commit configuration with ruff and black
- Fix lint CI job to use uv tool install instead of sync
- Add essential LlamaIndex dependencies to leann-core

Co-Authored-By: Yichuan Wang <73766326+yichuan-w@users.noreply.github.com>
2025-07-27 01:24:24 -07:00
Andy Lee
b3e9ee96fa fix: resolve all ruff linting errors and add lint CI check
- Fix ambiguous fullwidth characters (commas, parentheses) in strings and comments
- Replace Chinese comments with English equivalents
- Fix unused imports with proper noqa annotations for intentional imports
- Fix bare except clauses with specific exception types
- Fix redefined variables and undefined names
- Add ruff noqa annotations for generated protobuf files
- Add lint and format check to GitHub Actions CI pipeline
2025-07-26 22:38:13 -07:00
GitHub Actions
8375f601ba chore: release v0.1.13 2025-07-27 01:08:17 +00:00
yichuan520030910320
c87c0fe662 update colab install & fix colab path 2025-07-26 18:07:31 -07:00
GitHub Actions
802020cb41 chore: release v0.1.12 2025-07-26 23:35:28 +00:00
yichuan520030910320
cdb92f7cf4 update pytoml version && fix colab env && fix pdf extract in pip 2025-07-26 16:33:13 -07:00
yichuan520030910320
dc69bdec00 Merge branch 'main' of https://github.com/yichuan-w/LEANN 2025-07-25 17:54:43 -07:00
yichuan520030910320
98073e9868 update missing pkg 2025-07-25 17:54:21 -07:00
GitHub Actions
cf2ef48967 chore: release v0.1.11 2025-07-26 00:12:37 +00:00
yichuan520030910320
0692bbf7a2 change workflow 2025-07-25 17:11:56 -07:00
GitHub Actions
52584a171f chore: release v0.1.10 2025-07-25 23:12:16 +00:00
Andy Lee
efd6b5324b fix: add protobuf as a dependency for DiskANN backend
- Fixes 'No module named google' error when starting DiskANN embedding server
- Prevents users from having to manually install protobuf
2025-07-25 16:10:25 -07:00
Andy Lee
2baaa4549b fix: handle relative paths in HNSW embedding server metadata
- Convert relative paths to absolute paths based on metadata file location
- Fixes FileNotFoundError when starting embedding server
- Resolves issue with passages file not found in different working directories
2025-07-25 16:09:53 -07:00
GitHub Actions
75ddcd6158 chore: release v0.1.9 2025-07-25 20:04:42 +00:00
Andy Lee
2a4df11f5c fix: absolute path for passages 2025-07-25 11:59:30 -07:00
yichuan520030910320
cd8b970eff Merge branch 'main' of https://github.com/yichuan-w/LEANN 2025-07-25 01:45:57 -07:00
yichuan520030910320
52153bbb69 update faiss compare 2025-07-25 01:45:50 -07:00
GitHub Actions
e1ae087207 chore: release v0.1.8 2025-07-25 08:24:40 +00:00
Andy Lee
48c5e12ac1 fix: use absolute path for passages_file to prevent FileNotFoundError
When embedding server is launched as a subprocess, it may run in a different
working directory. Using absolute paths ensures the server can always find
the metadata file regardless of where it's launched from.
2025-07-25 01:23:47 -07:00
Andy Lee
673fd9b7cd fix: upgrade to actions v4 and handle manylinux2014 compatibility
- Upgrade all GitHub Actions to v4 (v3 is deprecated)
- Use manual git checkout in manylinux2014 containers to avoid Node.js issues
- Update artifact naming to ensure uniqueness (required by v4)
- Add fail-fast: false to build strategies
- This maintains manylinux2014 compatibility while using latest actions
2025-07-25 00:20:21 -07:00
yichuan520030910320
800d4cf111 Merge branch 'main' of https://github.com/yichuan-w/LEANN 2025-07-25 00:12:47 -07:00
yichuan520030910320
b6d43f5fd9 add gif 2025-07-25 00:12:35 -07:00
GitHub Actions
e64b599276 chore: release v0.1.7 2025-07-25 04:47:57 +00:00
GitHub Actions
166986d5e6 chore: release v0.1.6 2025-07-25 04:30:07 +00:00
GitHub Actions
ed27a127d5 chore: release v0.1.5 2025-07-25 04:00:54 +00:00
GitHub Actions
9000a7083d chore: release v0.1.4 2025-07-25 02:23:36 +00:00
GitHub Actions
20f2aece08 chore: release v0.1.3 2025-07-25 02:05:11 +00:00
GitHub Actions
cea1f6f87c chore: release v0.1.2 2025-07-25 01:53:29 +00:00
Andy Lee
133e715832 fix: resolve CI issues and consolidate workflows
- Fix version dependencies: update backend packages to depend on leann-core==0.1.1
- Remove duplicate ci.yml workflow (keeping build-and-publish.yml as main CI)
- Update release-manual.yml to reference correct CI workflow name

This fixes the dependency resolution error and eliminates duplicate builds.
2025-07-24 17:20:58 -07:00
GitHub Actions
faf5ae3533 chore: release v0.1.1 2025-07-24 23:36:23 +00:00
Andy Lee
a44dccecac fix: make TestPyPI upload optional and non-blocking
- Add continue-on-error to TestPyPI step
- Check if TEST_PYPI_API_TOKEN exists before attempting upload
- Add graceful failure handling with clear messages
- Update docs to explain TestPyPI token configuration
- Clarify that TestPyPI testing is optional

Now the release won't fail if TestPyPI is not configured or upload fails
2025-07-24 16:02:07 -07:00
yichuan520030910320
de252fef31 [chat] update 30s example 2025-07-24 14:40:33 -07:00
Andy Lee
7add391b2c chore: build and package 2025-07-24 00:47:46 -07:00
yichuan520030910320
efd6373b32 [chat] update huggingface chat and make qwen no thinking 2025-07-24 00:11:42 -07:00
yichuan520030910320
88eca75917 fix readme 2025-07-23 18:22:10 -07:00
yichuan520030910320
42de27e16a Merge branch 'main' of https://github.com/yichuan-w/LEANN 2025-07-23 18:17:19 -07:00
yichuan520030910320
c083bda5b7 fix several bug 2025-07-23 18:17:11 -07:00
Andy Lee
e86da38726 fix: ollama hint for similar models 2025-07-23 15:45:10 -07:00
yichuan520030910320
851f0f04c3 fix some para 2025-07-23 01:46:34 -07:00
yichuan520030910320
5dd74982ba fix readme 2025-07-22 23:14:31 -07:00
Andy Lee
30e5f12616 docs: quick start 2025-07-22 22:33:04 -07:00
yichuan520030910320
aa9a14a917 make the email wonderful format 2025-07-22 21:41:58 -07:00
Andy Lee
43155d2811 fix: supress resources leak logs 2025-07-22 19:53:45 -07:00
Andy Lee
d3f85678ec perf: much faster loading and embedding serving 2025-07-22 19:38:22 -07:00
yichuan520030910320
2a96d05b21 upd readme 2025-07-22 17:06:33 -07:00
Andy Lee
8513471573 feat: make diskann runnable 2025-07-22 14:26:03 -07:00
Andy Lee
71e5f1774c docs: cli 2025-07-21 23:48:40 -07:00