This repository collects notebooks that explore Retrieval-Augmented Generation (RAG) for various applications, from introductory to advanced implementations. It includes a full_basic_rag.ipynb starter and notebooks on multi-querying, routing, query construction, indexing, and advanced retrieval. The materials cover environment setup, data loading, embedding generation, and vector store configuration with ChromaDB or Pinecone.
Directory
Search results
Published directory entries matching your search.
cutorch is a CUDA backend for Torch7. It provides a torch.CudaTensor type that behaves like torch.FloatTensor with operations on the GPU, along with other GPU tensor types limited to copying, conversion, and certain indexing and shaping operations. The cutorch module offers functions for setting the GPU, reading device properties and memory usage, and managing streams and seeds.
S2 is a Go library for spherical geometry. It provides representations of angles, intervals, latitude-longitude points and unit vectors, geometric shapes over the unit sphere such as caps, rectangles, polylines and polygons, a hierarchical cell decomposition, constructive operations and predicates, in-memory indexing, distance algorithms and snapping and simplification. Planar geometry and conversions to common GIS formats are outside its scope.
Kubeflow is described as the cloud native AI platform, composed of modular open source projects forming a Kubernetes-native stack for data and AI workloads. Its subprojects include Kubeflow Pipelines for building and deploying portable machine learning workflows, Kubeflow Trainer for distributed AI training and LLM fine-tuning, Kubeflow Katib for automated machine learning, Kubeflow Notebooks for interactive development environments, and Kubeflow Hub for indexing models and metadata. It is a Cloud Native Computing Foundation project.
LlamaIndex is a platform focused on AI agents for document optical character recognition (OCR) and workflow automation. Its LlamaParse tool is designed to process complex documents including those with tables, charts and images, and can perform parsing, extraction, splitting, classification and indexing of document content.
RasaGPT is described as a headless LLM chatbot platform built on top of Rasa and Langchain. It is boilerplate and a reference implementation of Rasa and Telegram utilizing an LLM library like Langchain for indexing, retrieval and context injection.
A Ruby implementation of the UEA-Lite stemmer for conservative stemming in search and indexing workloads. It normalizes suffixes with a rule set while avoiding aggressive stemming, removes possessive apostrophes, expands contractions by default and preserves tokens beginning with uppercase letters. The gem has no runtime dependencies.