Provides efficient nearest-neighbor search and vector indexing for ML workflows.
npx skills add https://github.com/ldzhhxx/Hermes_offline_v2 --skill faiss-ldzhhxx
Please help me install this Agent Skill. Skill: faiss Source: https://github.com/ldzhhxx/Hermes_offline_v2/tree/main/hermes-agent/optional-skills/mlops/faiss Command: npx skills add https://github.com/ldzhhxx/Hermes_offline_v2 --skill faiss-ldzhhxx
This Skill provides the FAISS library, a powerful tool for efficient similarity search and clustering of dense vectors. It addresses the need for fast, large-scale vector retrieval and k-NN search, ideal for high-performance applications with billions of vectors.
Install the faiss skill and use it to search for the k-nearest neighbors of a vector:
pip install faiss-cpu
import faiss
import numpy as np
d = 128 # Dimension
nb = 1000 # Number of vectors
vectors = np.random.random((nb, d)).astype('float32')
index = faiss.IndexFlatL2(d)
index.add(vectors)
k = 5 # Number of neighbors
query = np.random.random((1, d)).astype('float32')
distances, indices = index.search(query, k)
print(f"Nearest neighbors: {indices}")
print(f"Distances: {distances}")
High-intent search queries and answers about installing and using this skill.
Vector similarity search on large datasets is handled by indexing dense vectors with FAISS, enabling fast k-NN retrieval across billions of vectors. It supports various index types like Flat, IVF, and HNSW for different performance needs.
Yes, large-scale k-NN search supports GPU acceleration using faiss-gpu. This provides high throughput and low latency when querying dense vectors, significantly speeding up retrieval compared to CPU-based processing.
The best way to find k-nearest neighbors for dense vectors is using an indexed FAISS structure like IndexFlatL2. You add your numpy float32 vectors to the index and execute a search query to retrieve distances and indices.
Yes, you need numpy to perform similarity search without metadata because vector operations require numpy arrays formatted as float32. The FAISS library relies on numpy to structure dense vectors before indexing and searching.
FAISS supports different index types for vector retrieval, including Flat, IVF, and HNSW. These index index types allow you to balance search speed, accuracy, and memory usage based on your specific similarity search requirements.
JamesFincherPerform k-NN similarity search over dense float32 vectors using FAISS index structures.
Quill-AgentPerforms fast k-NN search and clustering on dense vectors using FAISS index types.
davpatel605-beepStore and retrieve vector embeddings for semantic search and document retrieval.
FalkorDBCreate and query HNSW vector indexes in FalkorDB graph databases.