This repository serves as an archive and documentation site for FOIA materials related to the MU5735 aircraft incident investigation. It contains PDF documents, email records, FDR data in CSV format, and visualizations of flight data. The repository does not perform active data processing but preserves and presents source data for analysis. The most notable technical component is an interactive FDR data visualization built with vanilla JavaScript that reads CSV files directly and provides filtering, zooming, and multi-chart display capabilities. This visualization component demonstrates good separation of concerns and could be reused for similar time-series CSV datasets with minimal adaptation.
This repository is a curated collection of accessibility resources organized as an 'awesome list'. It contains no executable code or data processing pipelines, but rather serves as a reference directory linking to external accessibility tools, articles, guidelines, and communities. The content is structured into topic-specific markdown files (e.g., tools.md, blogs.md, laws.md) that catalog resources with metadata like names, descriptions, and languages. The project relies on community contributions following strict guidelines to maintain alphabetical order, avoid duplicates, and ensure quality. A GitHub Actions workflow automatically validates markdown links on every push to combat link rot, demonstrating a commitment to data integrity. As a pure data repository, its value lies in the organized aggregation of third-party resources rather than any internal transformation or analysis.
This repository is a curated list of awesome BASIC dialects, IDEs, tools, and tutorials. It serves as a reference for developers interested in the BASIC programming language ecosystem, providing organized links to resources categorized by dialects, development tools, miscellaneous projects, and learning tutorials. The repository consists primarily of a single README.md file that contains the curated list, supplemented by CONTRIBUTING.md guidelines for maintaining the list's quality and consistency, and a standard LICENSE file. There is no executable code or data processing pipeline; the value lies in the human-curated collection of links. The CONTRIBUTING.md file stands out as a reusable asset, offering a template for contribution guidelines that can be adapted to other awesome lists or curated repositories. It emphasizes consistent formatting, quality checks, and clear contribution procedures, demonstrating practical data stewardship practices for community-maintained resource lists.
This repository is a curated database of Bluetooth Assigned Numbers, providing JSON-formatted data for Company IDs, Service UUIDs, Characteristic UUIDs, Descriptor UUIDs, and GAP Appearance values. It serves as a single source of truth for Bluetooth numeric identifiers, enabling developers to look up human-readable names and descriptions for these values. The project includes validation and verification scripts to ensure data integrity, and update scripts to maintain synchronization with the official Bluetooth SIG assignments. While the core value lies in the JSON data files, the repository also provides reusable patterns for data validation, duplicate detection, and external data transformation.
Python-100-Days (Python 100倩δ»ζ°ζε°ε€§εΈ) is one of the most widely referenced Chinese-language Python learning curricula on GitHub (~186k stars). Despite its scale, it is fundamentally an educational artifact rather than a deployable software system: the dominant content is a 100-day sequence of Markdown tutorials (Day01-100) augmented with teaching-oriented code examples, SQL schema scripts, and Jupyter notebooks. The closest software classification is data_research, driven by the substantial data-analysis track β Day66-80 covers NumPy, pandas, and Matplotlib against real bundled datasets (scraped BOSSη΄θ job postings across nine Chinese cities, 2020 sales data, 2023 Beijing hukou data), and Day81-90 covers scikit-learn machine learning with classic datasets (iris, Titanic, MPG, Boston housing). The repo's character is hybrid: it also carries a Django web-development track (Day46-60), a web-scraping track with Scrapy/Selenium (Day61-65), and database/DevOps chapters (Day36-45, Day91-100), but none of these constitute a runnable production application β they are teaching material. Ground truth from the code: the only real end-to-end 'data pipeline' is the job-market dataset pipeline. Raw CSVs under Day66-80/code/res/jobs/ contain scraped listings from zhipin.com with a consistent schema (company_name, uri, salary, site, year, edu, job_name, city, pos_count); the notebook workflow produces a cleaned artifact (cleaned_jobs-checkpoint.csv) with salary ranges parsed into salary_lower/salary_upper bounds. This demonstrates a reusable salary-parsing and job-market normalization pattern. The machine-learning chapters are textbook-style reproductions of classical algorithms for pedagogy, not original research, and the scraping data is a timestamp-less snapshot. Maintenance is active: the author continues to revise the curriculum (a 2025 version reset of the introductory series) and the last commit is recent. Technical drift is medium β older chapters reference Django 2.2, Vue 2.6.10, ECharts 4.2.1, and Python 3.7-era idioms, while some third-party API examples (TianAPI, chaojiying captcha OCR, SMS gateways) depend on external services that may have changed. Reproducibility is partial: dataset files needed by notebooks are checked in, but the main curriculum has no pinned dependency manifest (only the opencourse subproject has requirements.txt), no conda environment file, and the scraped data has no crawl timestamp.
This repository is a comprehensive educational cookbook for building with Anthropic's Claude API and Agent SDK. It contains over 60 Jupyter notebooks demonstrating capabilities such as classification, retrieval-augmented generation (RAG), summarization, text-to-SQL, knowledge graph construction, content moderation, tool use, multimodal processing, extended thinking, and agent orchestration. The repo also includes production-oriented hosting patterns (Docker, Kubernetes, Modal, Cloudflare Workers) for deploying agents, and extensive evaluation frameworks using promptfoo with custom Python evaluators (BLEU, ROUGE, LLM-as-judge). While primarily a tutorial collection, it exhibits hybrid characteristics with reusable infrastructure code for agent hosting, content moderation pipelines, and notebook validation. The codebase is actively maintained with CI/CD workflows, but relies on beta APIs and hardcoded model names that introduce technical drift. Data assets are synthetic but realistic, enabling reproducible demonstrations of Claude's capabilities.
This repository is a research evaluation harness for Visual Place Recognition (VPR) models. It wraps a wide range of state-of-the-art VPR methods behind a common CLI, allowing users to run any supported model on standard VPR datasets (e.g., St Lucia) or a bundled toy dataset. The main entry point is test_dataset.py, which loads database/query image folders, extracts descriptors using the selected model, computes recall@N metrics, and saves logs and optional visualizations. The repo is designed to work with the companion VPR-datasets-downloader to standardize dataset format. Although it is a CLI tool, its primary purpose is research evaluation, so it is classified as data_research.
This repository is a comprehensive library of machine-readable taxonomies (machine tags) for threat intelligence and information sharing, primarily designed for the MISP platform. It contains over 200 taxonomy definitions in a standardized JSON format, each defining a namespace with predicates and values for tagging events, attributes, and indicators. The taxonomies cover a wide range of domains including cyber incident classification, malware analysis, information handling policies (TLP, PAP, IEP), regulatory compliance (GDPR, NIS2, EU AI Act), attack modeling (kill chain, Diamond Model), and specialized areas like disinformation, AI safety, and drone classification. While the core content is data, the repository also includes a set of reusable Python tools for validation, generation, and conversion of taxonomies, as well as CI/CD pipelines and a JSON schema. The hybrid nature is evident: it is primarily a data library but with significant application-like tooling for maintaining and extending the taxonomy collection.
This repository serves as Flashbots' public research hub, coordinating MEV-related research proposals (FRPs) and maintaining a curated publication index. The codebase consists primarily of structured markdown files for research proposals (in various stages: active, completed, stagnant) and a Python script that regenerates publication listings in JSON, CSV, and Markdown from a hardcoded list. The `docs/` folder contains a static HTML site with search and filtering capabilities for publications. There is no application code, data pipeline, or notebook analysis; the repository is purely a documentation and process coordination tool for blockchain MEV research. While not a typical data research project, it effectively manages metadata and provides a clean, reproducible publication build process.
This is the official specification repository for the General Transit Feed Specification (GTFS) and GTFS Realtime, hosted by Google and maintained by the community. It contains extensive documentation defining the structure, fields, and semantics of static transit schedule data (CSV-based) and real-time updates (Protocol Buffers schema). The repository includes governance documents, best practices, example feeds (e.g., sample-feed-1.zip), and asciipb examples. There is no application code, data pipelines, or notebooks; it is purely a specification standard. It serves as the authoritative reference for transit agencies (producers) and application developers (consumers) building interoperable transit tools.
This repository provides a collection of Python scripts for processing Pushshift Reddit dump files, which are zstandard-compressed ndjson files. The tools enable extraction, transformation, and loading of Reddit data, including splitting files by time or subreddit, merging data from multiple sources, backfilling missing data via APIs, and performing various analyses such as counting words, subreddits, and overlapping users. The codebase includes reusable utilities for reading compressed files and handling Reddit-specific ID formats, as well as sophisticated merge logic for combining data from different ingest sources.
This repository is a comprehensive collection of algorithm tutorials and solutions primarily based on LeetCode problems. It contains over 60 original articles covering various algorithm techniques and data structures, with solutions implemented in multiple programming languages including Java, Python, C++, JavaScript, and Go. The content is organized into thematic series covering dynamic programming, data structures, algorithm thinking patterns, and high-frequency interview questions. The repository serves as an educational resource for algorithm learning, providing detailed explanations of problem-solving frameworks and reusable code patterns. Each article typically includes problem descriptions, solution approaches with step-by-step reasoning, complete code implementations, complexity analysis, and links to related practice problems. The content emphasizes teaching algorithmic thinking patterns rather than just providing solutions. The codebase demonstrates solid educational value with well-structured explanations and reusable algorithm templates. However, as a documentation and education-focused project rather than a software application, it lacks traditional software engineering elements like dependency management, testing frameworks, or reproducible execution environments. The code snippets are designed for algorithm learning rather than production use.
This repository is a curated Markdown knowledge base β an "awesome list" β cataloging 300+ externally hosted tools, articles, Twitter threads, dashboards, and educational resources for on-chain cryptocurrency investigations, blockchain forensics, and crypto OSINT. It is classified as data_research only as the closest available fit; in reality it is neither a software application nor a data pipeline. The repository contains zero source code, no ETL scripts, no notebooks, no APIs, and no tests. The entire substance lives in README.md, organized into sections: data-analysis tools, visualization tools, clusterizers (AML clustering), hacks databases, VR/3D visualizations, Monero resources, a knowledge hub of methodology articles, and a list of investigators to follow.
This repository implements a personal knowledge management system centered around LLM-powered content processing. It follows a four-layer architecture: raw/ for immutable source materials, wiki/ for LLM-generated structured knowledge (summaries, concepts, indexes), brainstorming/ for intermediate outputs like chat logs and health reports, and artifacts/ for user-created work. The system processes various input formats (articles, books, PDFs, Facebook exports) through user-triggered commands like /compile and /convert-to-md, which leverage external tools (pandoc, pymupdf) and LLMs to extract insights and build interconnected knowledge networks. The core innovation lies in separating raw consumption from synthesized understanding, allowing users to accumulate source material without immediate processing, then periodically run LLM-driven synthesis to surface patterns across documents. Key workflows include extracting concepts that appear in multiple sources, maintaining bidirectional links between related ideas, and preserving tension between personal experience and external research in concept notes. The system emphasizes provenance tracking, with every synthesized piece linking back to its source via wikilinks. While the system relies on external LLMs and tools for its core functionality, it provides well-defined, reusable components for common data preparation tasks. The Facebook JSON-to-Markdown converter stands out as a robust, standalone script handling encoding normalization, media extraction, and frontmatter generation. The overall design demonstrates thoughtful consideration of knowledge work friction points, though reproducibility is hampered by undocumented environment dependencies and lack of sample data.
This is a research-grade PyTorch implementation of Danijar Hafner's Dreamer v4 model-based reinforcement learning algorithm. The core artifact is the `dreamer4` package, dominated by a monolithic `dreamer4/dreamer4.py` (~3000 lines) that implements `VideoTokenizer`/`SpaceTimeTokenizer`, `DynamicsWorldModel`, `AudioTokenizer`, and `StateTokenizer`. These classes build causal space-time transformers with separate space/time attention blocks, latent flow-matching diffusion with shortcut consistency, BYOL-style tokenizer pre-training, PoPE attention, and a large zoo of experimental loss terms (latent autoregressive loss, decorrelation, orthogonality, SigReg, self-flow distillation, shortcut loss). `dreamer4/trainers.py` provides training loops (VideoTokenizerTrainer, BehaviorCloneTrainer, DreamTrainer, SimTrainer) and `dreamer4/cli.py` exposes CLI entry points for training and serving the world model. The repository is hybrid: primarily research code, but packaged and tested like a library. It ships a pyproject.toml with pinned dependencies, a parametrized test suite (tests/test_dreamer.py, test_e2e.py, test_env_wrapper.py), a browser-based Snake environment with a Flask-style server (`dreamer4/web_env/`) used for human demonstration collection and replay-buffer inspection, and environment wrappers in `dreamer4/env.py` that record agent experience to folders or memmap replay buffers. Root-level scripts are the actual experiment entry points, targeting standard benchmarks: HalfCheetah (imagination RL), CartPole (dynamics + RL), MovingMNIST (tokenizer + dynamics), and Snake (PPO with dynamics model). Craftsmanship is above typical research code: clean package layout, pytest coverage across many hyperparameter combinations, self-contained utilities (LossNormalizer, tensor pack/unpack helpers, with_seed decorator, memmap replay-buffer helpers), and HuggingFace Accelerate integration for distributed/mixed-precision training. However, the single-file core, 100+ hyperparameter flags on trainers, scattered hard-coded constants, sparse docstrings, and reliance on volatile niche packages (x-transformers, h-net-dynamic-chunking, PoPE_pytorch, adam-atan2-pytorch) make long-term stability and code reuse risky. The repository is recently committed and appears maintained, but it is explicitly beta-status experimental research code, not a production system.
This repository is a curated 'awesome list' of research resources on multimodal learning in medical imaging. It contains no application code, tests, or data pipelines; its sole substantive file is a large README.md that catalogs papers, datasets, benchmarks, and code repositories. Topics span medical vision-language models, report generation, visual question answering, retrieval-augmented generation (RAG), and multimodal LLMs. The list is organized by task and includes links to arXiv PDFs, GitHub code, and dataset sources, with badges indicating publication venues.
The kjhealy/socviz repository is an R package that provides curated datasets and utility functions for teaching data visualization using ggplot2, primarily as a companion to the book "Data Visualization: A Practical Introduction" (2nd edition, 2026). It bundles approximately 30 social science datasets sourced from US Census Bureau data (county demographics, poverty), the General Social Survey, US election results (historic through 2024), OECD economic indicators, FRED time series, opiate mortality, and others. These datasets are clean, well-documented, and compressed for efficient loading via `data()`. The package also includes small utility functions (e.g., `theme_socviz`, `%notin%`, `round_df`) that support common data visualization workflows. While the primary value lies in the ready-to-use datasets rather than complex transformations, the package is properly structured with roxygen2 documentation, pkgdown site, and CRAN CI checks. It is a solid resource for educators and students learning R graphics, though reproducibility of the raw data pipelines depends on external sources that may change over time.
This repository (OSGL) is a curated catalog of openβsource games, game engines, frameworks and tools. It ingests metadata from external APIs (GitHub, GitLab, SourceForge, LibreGameWiki) and transforms them into a collection of structured Markdown files (over 1800 entries). A static site generator (Python + Jinja2 + Bulma) produces a browsable website with alphabetical, genre, platform and inspiration indices. The project provides maintenance scripts (Qt UI, batch validation, screenshot management) and synchronisation modules for keeping the catalog upβtoβdate. It is primarily a data curation and publishing project rather than a runnable application, but its Python tooling demonstrates good engineering for a data pipeline.
This repository is a research codebase for implementing and experimenting with dense retrieval techniques. It provides implementations of multiple advanced retrieval models including DPR, CITADEL, COIL, ColBERT, SPLADE, and their variants, along with data augmentation strategies like DRAGON and DRAMA. The codebase enables researchers to train, evaluate, and compare different retrieval approaches on standard benchmarks like Natural Questions, MS MARCO, and conversational datasets. It features a modular architecture with separate components for data processing, model definitions, training tasks, and evaluation, configured through Hydra for easy experimentation.
This repository curates and distributes Maxar Open Data satellite imagery for disaster response events. It provides a structured catalog of pre- and post-event high-resolution imagery (WorldView, GeoEye) in multiple formats: GeoJSON footprints, union geometries, MosaicJSON tile indices, and TSV metadata tables. The core pipeline (`maxar_data_catalog.py`) fetches the official Maxar STAC catalog, iterates over event collections, and generates these derived files, which are then committed to the repository. A Streamlit web app (`streamlit_app.py`) offers interactive browsing of footprints and mosaics. The repository is primarily a data asset with a supporting ETL pipeline and a thin frontend, making it a hybrid of data curation and a lightweight application. The data is organized by disaster event (e.g., hurricanes, earthquakes, floods) and includes rich acquisition metadata (platform, GSD, cloud cover, viewing angles) suitable for geospatial analysis and machine learning.