Directory
Published directory entries matching your search.
Haystack is an open-source AI orchestration framework for building production-ready LLM applications in Python. It provides modular pipelines and agent workflows with explicit control over retrieval, routing, memory and generation, and is intended for RAG systems, multimodal applications, semantic search, question answering and autonomous agents. Installation is available via pip.
Hydrosphere Mist is a serverless proxy for Spark clusters. It offers a functional programming framework and deployment model for Spark applications, with a REST HTTP and messaging API for Scala and Python Spark jobs. Sessions are managed across EMR, Hortonworks, Cloudera, DC/OS and vanilla Spark clusters.
Hypertunity is a lightweight, high-level Python library for black-box hyperparameter optimisation. It supports Bayesian optimisation by wrapping GPyOpt, objective function evaluation through a scheduler that is also compatible with Slurm, and real-time visualisation of results in Tensorboard via the HParams plugin. A Trial wrapper class allows parallel optimisation without manual scheduling, optimiser updates and logging.
A python library for doing approximate and phonetic matching of strings.
Jieba is a Chinese text segmentation module for Python. It supports four segmentation modes, including precise, full, search-engine and a paddle mode using PaddlePaddle, and offers keyword extraction based on TF-IDF and TextRank.
Kaveen Kumarasinghe is a software engineer in Seattle working at Stripe and previously part-time at OpenAI, building open-source tools and AI experiments. The site lists projects including GPTDiscord, an open-source Python integration that brings GPT conversations, image generation and document questions into Discord.
Keras GPT Copilot is a Python package that integrates an LLM copilot within the Keras model development workflow, generating feedback from model configuration, optimizer details and experiment results. It interacts with OpenAI's GPT models and offers a prompt-only mode for compatibility with other large language models.
KerasTuner is a hyperparameter optimization framework for Keras. It provides a define-by-run syntax for configuring search spaces and includes Bayesian Optimization, Hyperband and Random Search algorithms. It requires Python 3.8+ and TensorFlow 2.0+.
KoNLPy (pronounced "ko en el PIE") is a Python package for natural language processing of the Korean language. It is open source software released under GPL v3 or above, and users are asked to acknowledge its use with a citation.
A Python framework for building LLM-powered applications through a multi-agent paradigm. Developers set up agents, equip them with optional components such as LLMs, vector stores and tools, assign tasks, and have them exchange messages to solve problems. It works with practically any LLM and does not use LangChain.
Analyzes structured data and helps produce queries, insights or visualizations.
Lifelines is a complete survival analysis library, written in pure Python.
Lightwood is an AutoML framework that generates and customizes machine learning pipelines using a declarative JSON-AI syntax. It handles data types such as numbers, dates, categories, tags, text, arrays and multimedia formats, and includes a time-series mode. Users can override defaults or replace pipeline steps, and the framework generates Python code for the pipeline.
Little Ball of Fur is a graph sampling extension library for Python and a graph sampling extension library for NetworKit and NetworkX. It consists of methods that can sample from graph structured data, covering vertex, edge, and exploration sampling techniques with a unified application public interface.
A self-hosted proxy providing OpenAI and Anthropic compatible API endpoints for multiple LLM providers, so existing applications can use any supported provider through a single API key. It includes a FastAPI proxy application and a reusable Python resilience library for API key management, rotation, and failover.
LocalGPT is an on-premise document intelligence platform that lets users chat with their documents on their local device using GPT models. It supports PDF, DOCX, HTML, Markdown, and TXT files and combines dense vector search with LanceDB full-text search and a cross-encoder reranker. It runs with a plain Python RAG core and offers generation models via Ollama.
m2cgen is a tool that transforms trained machine learning models into native code across a wide range of programming languages including Java, C, Python, Go, JavaScript, C#, R, PHP, Dart, Haskell, Ruby, F#, and Rust, with zero dependencies. Its codebase is hosted in the BayesWitnesses GitHub repository, which includes navigation for installation, development guidance, supported languages, and supported model types.
Machine Learning Zoomcamp is a free, hands-on course on building, evaluating, and deploying machine learning systems. It covers the path from framing a problem to preparing data, training and evaluating models, and deploying them, using Python, scikit-learn, TensorFlow, PyTorch, FastAPI, Docker, Kubernetes and AWS Lambda. No prior machine learning experience is required.
The Materials Data Facility (MDF) is a service for publishing, discovering and accessing materials science datasets. It supports minting DOIs for datasets and offers curated collections covering simulations, experimental data and machine-learning-ready data. Data can be uploaded from local storage, Globus endpoints or Google Drive, and datasets can be loaded with a Python package.
Matt Harrison is a personal blog authored by Matt Harrison. The blog covers topics including Amazon book piracy, the Effective Pandas book, Python and data science courses, and general software development content.
Mengram is an open-source project that provides human-like memory capabilities for AI agents. It supports semantic, episodic, and procedural memory types, with experience-driven procedures designed to learn from failures. The project offers a free API, Python and JavaScript SDKs, and integrations with LangChain, CrewAI, and OpenClaw.
A Python module for metric learning.
Metric Learning Algorithms in Python.
Repository containing the source code of the Python package loralib and examples of integrating it with PyTorch models, such as those in Hugging Face. LoRA learns pairs of rank-decomposition matrices while freezing the original weights, reducing the number of trainable parameters. PyTorch is the only supported framework for now.
