mirror of https://github.com/ruvnet/RuVector.git synced 2026-05-24 05:43:58 +00:00

History

rUv 062130348d feat(postgres): Add 53 SQL function definitions for all advanced modules Enable all advanced PostgreSQL extension functions by adding their SQL definitions to the extension file. This exposes all Rust #[pg_extern] functions to PostgreSQL. ## New SQL Functions (53 total) ### Hyperbolic Geometry (8 functions) - ruvector_poincare_distance, ruvector_lorentz_distance - ruvector_mobius_add, ruvector_exp_map, ruvector_log_map - ruvector_poincare_to_lorentz, ruvector_lorentz_to_poincare - ruvector_minkowski_dot ### Sparse Vectors (14 functions) - ruvector_sparse_create, ruvector_sparse_from_dense - ruvector_sparse_dot, ruvector_sparse_cosine, ruvector_sparse_l2_distance - ruvector_sparse_add, ruvector_sparse_scale, ruvector_sparse_to_dense - ruvector_sparse_nnz, ruvector_sparse_dim - ruvector_bm25_score, ruvector_tf_idf, ruvector_sparse_normalize - ruvector_sparse_topk ### GNN - Graph Neural Networks (5 functions) - ruvector_gnn_gcn_layer, ruvector_gnn_graphsage_layer - ruvector_gnn_gat_layer, ruvector_gnn_message_pass - ruvector_gnn_aggregate ### Routing/Agents - "Tiny Dancer" (11 functions) - ruvector_route_query, ruvector_route_with_context - ruvector_calculate_agent_affinity, ruvector_select_best_agent - ruvector_multi_agent_route, ruvector_create_agent_embedding - ruvector_get_routing_stats, ruvector_register_agent - ruvector_update_agent_performance, ruvector_adaptive_route - ruvector_fastgrnn_forward ### Learning/ReasoningBank (7 functions) - ruvector_record_trajectory, ruvector_get_verdict - ruvector_distill_memory, ruvector_adaptive_search - ruvector_learning_feedback, ruvector_get_learning_patterns - ruvector_optimize_search_params ### Graph/Cypher (8 functions) - ruvector_graph_create_node, ruvector_graph_create_edge - ruvector_graph_get_neighbors, ruvector_graph_shortest_path - ruvector_graph_pagerank, ruvector_cypher_query - ruvector_graph_traverse, ruvector_graph_similarity_search ## CLI Updates - Enabled hyperbolic geometry commands in postgres-cli - Added vector distance and normalize commands - Enhanced client with connection pooling and retry logic 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>		2025-12-03 03:44:14 +00:00
..
src	feat(postgres): Add 53 SQL function definitions for all advanced modules	2025-12-03 03:44:14 +00:00
package.json	fix(postgres): Resolve compilation errors and Docker build issues	2025-12-03 01:17:50 +00:00
README.md	fix(postgres): Resolve compilation errors and Docker build issues	2025-12-03 01:17:50 +00:00
tsconfig.json	fix(postgres): Resolve compilation errors and Docker build issues	2025-12-03 01:17:50 +00:00

README.md

@ruvector/postgres-cli

Command-line interface for the RuVector PostgreSQL extension - an advanced AI vector database.

Installation

npm install -g @ruvector/postgres-cli

Quick Start

# Connect to your PostgreSQL database with RuVector extension
ruvector-pg -c "postgresql://user:pass@localhost:5432/mydb" info

# Install the extension
ruvector-pg install

# Create a vector table
ruvector-pg vector create embeddings --dim 384 --index hnsw

# Search vectors
ruvector-pg vector search embeddings --text "hello world" --top-k 10

Commands

Vector Operations

# Create vector table with HNSW index
ruvector-pg vector create <name> --dim <dimensions> --index <hnsw|ivfflat>

# Insert vectors from JSON file
ruvector-pg vector insert <table> --file vectors.json

# Search for similar vectors
ruvector-pg vector search <table> --query "[0.1, 0.2, ...]" --top-k 10 --metric cosine

Attention Mechanisms

# Compute attention
ruvector-pg attention compute --query "[...]" --keys "[[...]]" --values "[[...]]" --type scaled_dot

# List available attention types
ruvector-pg attention list-types

Graph Neural Networks

# Create GNN layer
ruvector-pg gnn create my_layer --type gcn --input-dim 384 --output-dim 128

# Forward pass
ruvector-pg gnn forward my_layer --features features.json --edges edges.json

Graph & Cypher

# Execute Cypher query
ruvector-pg graph query "MATCH (n:Person) RETURN n"

# Create node
ruvector-pg graph create-node --labels "Person,Developer" --properties '{"name": "Alice"}'

# Traverse graph
ruvector-pg graph traverse --start node123 --depth 3 --type bfs

Self-Learning

# Train from trajectories
ruvector-pg learning train --file trajectories.json --epochs 10

# Make prediction
ruvector-pg learning predict --input "[0.1, 0.2, ...]"

Benchmarking

# Run benchmarks
ruvector-pg bench run --type all --size 10000 --dim 384

# Generate report
ruvector-pg bench report --format table

Global Options

-c, --connection <string> - PostgreSQL connection string (default: postgresql://localhost:5432/ruvector)
-v, --verbose - Enable verbose output

Features

Vector Search: HNSW and IVFFlat indexes with cosine, L2, and inner product metrics
39 Attention Mechanisms: Scaled dot-product, multi-head, flash, sparse, and more
Graph Neural Networks: GCN, GraphSAGE, GAT, GIN layers
Graph Operations: Cypher queries, BFS/DFS traversal
Self-Learning: ReasoningBank-based trajectory learning
Hyperbolic Embeddings: Poincaré and Lorentz models
Sparse Vectors: BM25 and SPLADE for hybrid search

License

MIT