BREAKING CHANGE: Complete migration from TensorFlow.js to Transformers.js for embedding generation
This is a major architectural change that replaces TensorFlow.js (USE model) with Transformers.js (all-MiniLM-L6-v2) for significantly improved performance and reduced complexity.
Key Changes:
- Replace TensorFlow.js Universal Sentence Encoder with Transformers.js all-MiniLM-L6-v2
- Reduce model size from 525MB to 87MB (83% reduction)
- Reduce embedding dimensions from 512 to 384 (faster distance calculations)
- Remove TensorFlow.js Float32Array patching (caused ONNX conflicts)
- Implement smart bundled model detection for offline operation
- Add explicit model download script for Docker deployments
- Remove complex environment variables in favor of simple configuration
- Update all distance functions to use optimized pure JavaScript
- Remove TensorFlow-specific utilities and type definitions
Performance Improvements:
- Model loading: 5x faster (87MB vs 525MB)
- Memory usage: 75% reduction (~200-400MB vs ~1.5GB)
- Distance calculations: Faster pure JS vs GPU overhead for small vectors
- Cold start performance: Significantly improved
Files Changed:
- Updated package.json: New dependencies, simplified scripts
- Rewrote src/utils/embedding.ts: Complete Transformers.js implementation
- Updated src/utils/distance.ts: Optimized JavaScript distance functions
- Simplified src/setup.ts: Removed TensorFlow-specific patching
- Simplified src/utils/textEncoding.ts: Only Node.js TextEncoder/Decoder patches
- Deleted src/utils/robustModelLoader.ts: TensorFlow-specific loader
- Deleted src/types/tensorflowTypes.ts: TensorFlow type definitions
- Added scripts/download-models.cjs: Docker-compatible model downloader
- Added comprehensive documentation: README.md, OFFLINE_MODELS.md, analysis docs
Testing:
- All 19 tests passing
- Removed test mocking in favor of real implementation testing
- Updated test environment for Transformers.js compatibility
- Performance tests validate improved efficiency
This migration resolves production issues with Docker egress limitations and provides a more robust, performant foundation for vector operations.
Implements Phase 1 and Phase 2 of distributed enhancements for horizontal scaling:
Phase 1 - Zero-Config Distributed Mode:
- Add DistributedConfigManager for shared S3 configuration coordination
- Implement explicit role configuration (reader/writer/hybrid) for safety
- Add instance registration with heartbeat and health monitoring
- Create hash-based partitioner for deterministic data distribution
Phase 2 - Intelligent Data Management:
- Add DomainDetector for automatic data categorization (medical, legal, product, etc.)
- Implement domain-aware search filtering for improved relevance
- Create role-based operational modes with specific optimizations
- Add HealthMonitor for comprehensive metrics tracking
Key Features:
- Multi-writer support with consistent hash partitioning
- Reader instances optimize for 80% cache utilization
- Writer instances optimize for batched writes
- Automatic domain detection and tagging
- Real-time health monitoring across all instances
- Cross-platform crypto utilities for browser compatibility
Safety Improvements:
- Require explicit role configuration (no automatic assignment)
- Validate role compatibility on startup
- Track instance health and performance metrics
Testing:
- Add comprehensive test suite for distributed features
- All 25 distributed tests passing
- Fixed domain filtering in search functionality
Documentation:
- Update README with distributed mode highlights
- Add examples showing reader/writer setup
- Document new capabilities and benefits
🤖 Generated with Claude Code
https://claude.ai/code
Co-Authored-By: Claude <noreply@anthropic.com>
- Introduced `PaginationOptions`, `NounFilterOptions`, and `VerbFilterOptions` types for improved query flexibility in data retrieval operations.
- Added `getNouns` and `getVerbs` methods with pagination and filtering capabilities, replacing existing methods for broader use cases and scalability.
- Marked legacy methods (`getAllNouns`, `getAllVerbs`, `getVerbsBySource`, `getVerbsByTarget`, `getVerbsByType`) as deprecated, directing users to use new methods.
- Updated `coreTypes`, `memoryStorage`, and related modules to support new functionality, including cursor and offset-based pagination handling.
- Updated fallback logic for storage adapters, ensuring compatibility with non-paginated operations when required.
**Purpose**: Enhance scalability and query precision by implementing paginated and filtered retrieval of nouns and verbs, aligning query methods with modern requirements.
- Reformatted export statements across `index.ts` and related modules for consistent style, improving code readability and maintainability.
- Updated graph types in `graphTypes.ts` to include additional standardized noun and verb categories, enhancing the flexibility of the type system for graph modeling.
- Replaced `Place` with `Location` and merged similar types (e.g., `Group` into `Collection`) to eliminate redundancy in entity definitions.
- Expanded verb types in `VerbType` to cover more comprehensive use cases, including social, temporal, and ownership relationships.
**Purpose**: Streamline code structure with consistent export formatting, simplify type definitions, and enhance the type system for broader modeling capabilities.
- **Removed Files**:
- Deleted outdated statistics documentation files (`statistics.md`, `statistics-flush-solution.md`, `statistics-summary.md`) to clean up the repository and avoid confusion.
- **Added Standards**:
- Introduced `DOCUMENTATION_STANDARDS.md` to outline naming conventions and troubleshooting practices for more consistent and maintainable project documentation.
- **Tests**:
- Added a new test file `edge-cases.test.ts` to verify handling of edge cases, ensuring robust behavior against boundary values and invalid inputs.
**Purpose**: Cleans up deprecated documentation while introducing concrete standards for maintaining and updating documentation. Enhances test coverage for unusual or boundary inputs, improving overall system resilience.
- **Core**: Improved verb creation logic by adding `createdAt`, `updatedAt`, and `createdBy` attributes. These fields include timestamped metadata (`seconds`, `nanoseconds`) and source augmentation/service information for better tracking.
- **Storage**: Refactored `BaseStorage` methods to utilize internal variants (e.g., `saveVerb_internal`, `getNoun_internal`). Added support for new verb attributes while maintaining backward compatibility with existing data structures.
- **Tests**:
- Updated `s3-storage.test.ts` and `opfs-storage.test.ts` to validate changes in verb attributes such as timestamps and augmentation metadata.
- Added assertions for `createdAt`, `updatedAt`, and `createdBy` fields in test cases.
- **Cleanup**: Replaced ambiguous type aliases like `Edge` and `HNSWNode` with clearer equivalents (`Verb` and `HNSWNoun_internal`) for consistency across storage adapters.
**Purpose**: Enhance metadata tracking and standardize attribute handling across storage and core modules to ensure accurate and consistent data throughout the system.
- Added `getFile` method to `FileSystemHandle` interface for enhanced TypeScript compatibility with File System Access API.
- Improved maintainability by reformatting and aligning imports in `brainyData.ts`.
- Standardized spacing and indentation across structured comments and interface fields.
Purpose: Improve TypeScript support for file system operations and enhance code readability with consistent formatting and import management.
- Introduced `test-fallback-function.js` and `test-fallback-simple.js` to validate `executeInThread` fallback functionality with both named and anonymous compute-intensive functions.
- Added `test-tensorflow-textencoder.js` for TensorFlow.js and TextEncoder tests in a Node.js environment.
- Created `test-tensorflow-textencoder.html` for browser-based TensorFlow.js and TextEncoder tests.
- Implemented cross-environment test support in `cli-package/src/test-tensorflow-textencoder.ts` for CLI functionality.
- Enhanced `src/utils/embedding.ts`, `textEncoding.ts`, and `brainy-wrapper.js` to include updated global `TextEncoder` and `TextDecoder` utilities for compatibility and worker improvements.
- Standardized and expanded utility methods in `PlatformNode` for broader support, including `isFloat32Array` and `isTypedArray` checks.
- Updated Node.js requirement to `>= 24.4.0` across documentation and configuration files for compatibility improvements.
This update introduces comprehensive testing for fallback mechanisms, TensorFlow.js, and TextEncoder across multiple environments, ensuring robustness and compatibility.
- Removed outdated Node.js v24 compatibility patches for `TextEncoder` and `TextDecoder` in `cli-wrapper.js` and other modules since TensorFlow.js now supports Node.js v24+ natively.
- Introduced a utility module `src/utils/tensorflowUtils.ts` for TensorFlow.js compatibility, defining global `PlatformNode` and related utilities.
- Enhanced `fileSystemStorage.ts` with improved Node.js module loading mechanisms for better compatibility across environments.
- Simplified dependency requirements by reducing the minimum Node.js version to `>=23.0.0`.
- Updated `package.json`, `cli-package/package.json`, and `README.md` versions to `0.9.25`.
- Harmonized `cli-package` and main package dependencies to ensure version alignment.
- Improved global compatibility with TensorFlow.js in mixed Node.js environments.
This update modernizes TensorFlow.js integration, reduces maintenance efforts, and aligns compatibility with current Node.js and TensorFlow.js capabilities.
