Directory
Published directory entries matching your search.
Chunk Tuner is a tool that benchmarks chunking strategies for a RAG corpus and recommends a configuration. It benchmarks strategies such as fixed-token windows, recursive character splitting, semantic splitting and PDF structural chunking, scoring each against retrieval metrics. It is available as a Python library, a CLI and an MCP server.
ClearTK is a framework for developing statistical natural language processing components in Java, built on top of Apache UIMA. It is developed by the Center for Computational Language and Education Research (CLEAR) at the University of Colorado at Boulder. The project is built with Maven, and the repository includes sub-projects such as cleartk-feature and modules for several machine learning libraries.
ClojisR is a bridge between Clojure and R, described as "Clojure speaks statistics." It targets only JVM Clojure and supports a function-centric API for calling R functions on R objects from Clojure. It requires Linux, MacOS or WSL, JDK 1.8 or later, Clojure 1.9.0 or later, and the Rserve R package.
Clojure-openNLP is an open-source Clojure library that provides an interface to the Apache OpenNLP natural language processing toolkit. Hosted as a public GitHub repository, it includes usage instructions for the Leiningen build tool and sample usage from a Clojure REPL.
Awesome-Code-LLM is a curated list of language modeling research for code and other software engineering activities, plus related datasets. The GitHub repository is marked TMLR and links to an arXiv paper.
Open Code LLMs for Code Understanding and Generation.
Cog is an open-source tool that packages machine learning models in standard, production-ready containers. Users define the environment in a configuration file, and it generates a Docker image with Nvidia base images and dependency caching. Model inputs and outputs are defined in Python, from which Cog generates an OpenAPI schema and an HTTP inference server. Models can be deployed to own infrastructure or Replicate.
A lightweight tool designed to run frontier mixture of experts (MoE) models on existing user-owned hardware. It is built with pure C programming language, has zero external dependencies, and streams model experts directly from disk, described as a tiny engine capable of running immense models.
Colibri Core is an NLP tool and a C++ and Python library for working with basic linguistic constructions such as n-grams and skipgrams, i.e. patterns with one or more gaps of fixed or dynamic size. It counts and extracts patterns from large corpus data. The central tool, colibri-patternmodeller, builds, views, manipulates and queries pattern models.
Track your datasets, code changes, experimentation history, and models.
An open-source project that provides tools to debug, evaluate, and monitor large language model (LLM) applications, retrieval-augmented generation (RAG) systems, and agentic workflows, with features including comprehensive tracing, automated evaluations, and production-ready dashboards.
Plugin for CMS Adobe Experience Manager (AEM) or Composum Pages helping the editor to create / edit / translate texts.
OS-agnostic, system-level binary package manager and ecosystem.
DeepEval is an open-source LLM evaluation framework, described as similar to Pytest but specialized for unit testing LLM apps. It provides evaluation metrics such as G-Eval, task completion, answer relevancy and hallucination, which use LLM-as-a-judge and other NLP models running locally. It can evaluate LLM apps end-to-end, complete agent trajectories, and individual agent steps.
A curation of awesome tools, documents and projects about LLM Security.
Cornac is an open-source comparative framework for multimodal recommender systems, maintained by PreferredAI and hosted on GitHub. It provides tools for developing and evaluating recommender system models that process multiple types of input data.
CorpusLoaders is a Julia package that provides a variety of loaders for different natural language processing (NLP) corpora, maintained by JuliaText and hosted on GitHub. It simplifies accessing and loading diverse NLP datasets for research and development use cases.
Teaching LLMs memory management for unbounded context [[demo page]](https://memgpt.ai/) [[arxiv]](https://arxiv.org/abs/2310.08560).
Cybertron is a package in pure Go built upon spaGO that gives Go developers an interface for using NLP technologies with pre-trained Transformer models from the HuggingFace models repository. It is primarily aimed at running inference, with the possibility of adding fine-tuning in the future, and supports models including BERT, ELECTRA, BART, PEGASUS and MarianMT.
Julia package for Regularized Discriminant Analysis.
A GitHub repository from dair-ai containing explanations of key concepts in machine learning. The README lists papers with dates and descriptions, including the Transformer, Elmo, Marian MT, GPT, BERT and Transformer XL.
Darknet is an open source neural network framework written in C and CUDA that supports CPU and GPU computation. The repository also references YOLO object detection work, including YOLOv7 and Scaled-YOLOv4 papers and source code.
Data Driven Code is a GitHub repository containing code from a PyCon Canada 2016 session. The repository provides resources and examples related to data-driven programming techniques presented at the conference.
DEAP is an evolutionary computation framework for Python aimed at rapid prototyping and testing of ideas. It provides genetic algorithms over representations such as lists, arrays, sets, dictionaries, trees and Numpy arrays, genetic programming with prefix trees, evolution strategies including CMA-ES, multi-objective optimisation, and co-evolution of multiple populations. It supports parallelisation via multiprocessing and SCOOP and works out of the box with Python 3 since version 0.8.
