Milvus Adventures November 15, 2024

We have started a weekly Office Hours where you can schedule some time with our DevRel and Engineering team members to get our Milvus questions answered!

COMMUNITY

We had some fun at our meetups recently and there still is one more next week in San Francisco. But we always record a past sessions, so come check them out!

• Nov 19, 2024 | Unstructured Data San Francisco Meetup Register

  • Hakan Tekgul, Evaluating RAG pipelines built on unstructured data
  • James Luan, Dense Embeddings != Complete Search - a sneak peak of Milvus 2.5
  • Max Mathys, Gandalf: Insights from the World's Largest Red Team

Hot Topics

• What is GraphRAG? | Microsoft's GraphRAG advances traditional RAG architectures by leveraging graph databases to create more intelligent, context-aware search and retrieval systems for large language models.
• Similarity Metrics | Cosine similarity, a key metric for measuring how closely two vectors match, helps power modern search systems by finding the most relevant results based on semantic meaning rather than exact matches.
• Llama 3.2 function calling | Llama 3.2 introduces native function calling capabilities, allowing developers to define and trigger specific actions directly through the model's output—similar to GPT's function calling but now available in open source.
• text-embedding-3-small model | The text-embedding-3-small model delivers improved embedding quality at lower latency and cost, making it ideal for production vector search applications.
• Class activation maps | Class activation maps visualize which regions of an image influenced a neural network's decision, helping developers understand and debug computer vision models.
• Convolutional neural network | Convolutional neural networks process images by applying filters that detect patterns and features, forming the backbone of modern computer vision applications from facial recognition to medical imaging.
• knn algorithm in machine learning K-Nearest Neighbors (KNN) | The K-Nearest Neighbors (KNN) algorithm classifies data points by finding their closest matches in your training data, making it an intuitive yet powerful approach for tasks like recommendation systems and anomaly detection.
• Vector database benchmarks | VectorDBBench provides standardized performance metrics across vector databases, measuring critical factors like query latency, throughput, and recall to help teams make data-driven infrastructure decisions.

GITHUB REPOS

Milvus Milvus is an open-source vector database built to power embedding similarity search and AI applications.

Akcio: Enhancing LLM-Powered ChatBot with CVP Stack A full chatbot app all open-source for you to try out for your self!

GPT Cache. GPTCache is an open-source tool designed to improve the efficiency and speed of GPT-based applications by implementing a cache to store the responses generated by language models.

VectorDBBench. VectorDBBench is an open-source benchmarking tool to help you evaluate the performance of mainstream vector databases and cloud services with yoru specific use case.