ISSUE 461 · November 14, 2023Talks & ConferencesExciting Data Visualizations with ggplot2 ExtensionsGreat talk covering the universe of ggplot2 extensions. This is very well organized and includes lots of useful screenshots, examples and code snippets along the way. Follow the links to download slides and code. OpenAI DevDay Breakout SessionsVideos from OpenAI's first annual DevDay were released last week and now include the breakout sessions. The videos cover a variety of topics, including ways to maximize LLM performance, AI Ops, the business of AI, and more. Sessions range in length from 20-45 minutes each. Sponsored LinkSpending Hours Manually Pulling Data into Spreadsheets?You know the drill: constant downloads, uploads, and endless copy-paste. It's tedious, mistakes happen, and data becomes stale the moment it's entered. Coefficient connects to any data source in seconds, imports live data into your spreadsheet and effortlessly automates workflows. 20,000+ teams like Miro & Klaviyo generate real-time insights faster. Posts & TutorialsA Coder Considers the Waning Days of the CraftDo the rapidly evolving coding capabilities of LLMs mean that coders' days are numbered? Or will LLMs have the opposite effect and empower more people to develop useful coding skills? This is a well-written and nuanced look at the craft of coding in a world of LLMs. Maxar's Open Satellite FeedSatellite data can get expensive, fast, but there are options for freely available data that's associated with major crisis events. This post explores open satellite data from Maxar and shows how to access, wrangle, and analyze it using Python, GDAL and DuckDB. GraphCast: AI model for global weather forecastingWeather forecasts typically rely on carefully defined physics equations, which are then translated into computer algorithms and run on supercomputers. Graphcast is a new SOTA model that delivers 10-day predictions at unprecedented accuracy in under one minute. It's open-source and is currently the most accurate 10-day global weather forecasting system in the world. Tools & Codedata load tool (dlt)dlt is an open source library to automatically create datasets out of messy, unstructured data sources. You can use the library to move data from a variety of SQL and vector stores, data lakes, storage buckets, or local engines like DuckDB. It automates many cumbersome data engineering tasks and can by handled by anyone who knows Python. ResourcesData-Driven Science and EngineeringData-driven discovery is revolutionizing how we model, predict, and control complex systems. Now in its 2nd edition, this text explores the intersection of data-driven methods, machine learning, and classical fields of engineering mathematics and mathematical physics. Follow the links to download the PDF for free or to purchase it in print. The State of Open Data 2023The 2023 State of Open Data report summarizes the survey results from over 6000 researchers regarding their practices and challenges in sharing data. Generally, researchers want to share their data but for a variety of reasons, often don't. The report takes a more analytical approach than prior years and explores both ongoing and emerging issues, including needed support, getting credit, and the use of AI. OutlierSpain lives in flats: why our cities are built verticallyGreat example of a well-told data story. It's not about data or ways to work with data but the presentation here is awesome. You're getting the free version of Data Elixir. Upgrade to get monthly deep dives and access to Data Pilot, the AI data expert that's built from Data Elixir's archives. This month's deep dive is focused on software development for non-engineers. Upgrade now and level up your coding skills 👉
|
Поиск по этому блогу
Search1
123
вторник, 14 ноября 2023 г.
Data Elixir - Issue 461
Подписаться на:
Комментарии к сообщению (Atom)
Комментариев нет:
Отправить комментарий
Примечание. Отправлять комментарии могут только участники этого блога.