talk-data.com

We will explore tidymodels, a cohesive framework for modeling and machine learning in R.

Join us for a hands-on, beginner-friendly session where we’ll explore how to create interactive maps in R using Leaflet. We’ll start with a simple dataset, do some quick visualizations with ggplot2, and then build up to a map with interactive popups. No need to be a mapping expert — we’ll code together and experiment as we go. Bring your laptop with R and RStudio installed, and we’ll make some maps!

Note: this event was originally going to be about tidymodels, but we have had to make a change of plans, so the tidymodels workshop will be held at a later time.

External registration required at nyhackr.

This month we have Max Kuhn giving a talk about tidymodels.

After the talk we will give away both an in-person and a virtual ticket to The Data Science and AI Conference (the spiritual successor of the NY R Conference) taking place August 25-27. Members of this meetup can get a 20% discount on tickets with code nyhackr.

Thank you to NYU for hosting us.

Everybody attending must RSVP through the registration form at nyhackr. There is a charge for in-person and virtual tickets are free.

Space is extremely limited and in-person registration closes at 3 PM the day of the talk.

About the Talk: A lot! I’ll discuss updates to tidymodels related to: postprocessing, sparse data, multiparameter optimization, parallel processing, mirai, catboost, quantile regression, ordinal data, AI, and two new packages.

About Max: Max Kuhn is a Scientist at Posit, PBC (nee RStudio). He is working on improving R’s modeling capabilities and maintaining about 30 packages, including caret. He was a Senior Director of Nonclinical Statistics at Pfizer Global R&D in Connecticut. He has been applying models in the pharmaceutical and diagnostic industries for over 18 years. Max has a Ph.D. in Biostatistics. He and Kjell Johnson wrote the book Applied Predictive Modeling, which won the Ziegel award from the American Statistical Association, which recognizes the best book reviewed in Technometrics in 2015. He has co-written several other books: Feature Engineering and Selection, Tidy Models with R, and Applied Machine Learning for Tabular Data (in process).

The venue doors open at 6:30 PM America/New_York where we will continue enjoying pizza together (we encourage the virtual audience to have pizza as well). The talk, and livestream, begins at 7:00 PM America/New_York.

Remember, register at nyhackr.

A practical guide to reproducible and high impact mass spectrometry data analysis R Programming for Mass Spectrometry teaches a rigorous and detailed approach to analyzing mass spectrometry data using the R programming language. It emphasizes reproducible research practices and transparent data workflows and is designed for analytical chemists, biostatisticians, and data scientists working with mass spectrometry. Readers will find specific algorithms and reproducible examples that address common challenges in mass spectrometry alongside example code and outputs. Each chapter provides practical guidance on statistical summaries, spectral search, chromatographic data processing, and machine learning for mass spectrometry. Key topics include: Comprehensive data analysis using the Tidyverse in combination with Bioconductor, a widely used software project for the analysis of biological data Processing chromatographic peaks, peak detection, and quality control in mass spectrometry data Applying machine learning techniques, using Tidymodels for supervised and unsupervised learning, as well as for feature engineering and selection, providing modern approaches to data-driven insights Methods for producing reproducible, publication-ready reports and web pages using RMarkdown R Programming for Mass Spectrometry is an indispensable guide for researchers, instructors, and students. It provides modern tools and methodologies for comprehensive data analysis. With a companion website that includes code and example datasets, it serves as both a practical guide and a valuable resource for promoting reproducible research in mass spectrometry.

------ Welcome\, London R Users! ------ Welcome back! We can't wait to close the year with some great R talks :)

As before, the event will be brought to you by Datacove. We are a Data and Analytics Consultancy Team based in Brighton, working across Customer Analytics, Shiny and Reporting, Marketing Analytics and Process Automation.

We will be supported by the generous team at LSBU, organised by Valerio Ficcadenti from the Business School. Hosted by Jeremy Horne from Datacove.

------ Talk One ------ For our first talk, we are thrilled to be hosting Hannah Frick, Software Engineer at Posit.

As the author of Tidymodels, Hannah will be sharing 'What's New with Tidymodels?'. The `tidymodels` package in R is a collection of packages designed to streamline machine learning and modeling using a tidy and consistent framework. It integrates tools for model specification, data preprocessing, resampling, tuning, and evaluation. Key components include `parsnip` for fitting models with a unified syntax and `recipes` for flexible data preprocessing.

------ Talk Two ------ Datacove's own, Jeremy Horne, will be bringing us a quick but insightful talk to end the evening. As a man who has attended all 10 EARL Conferences - we're excited to hear his takeaways!

This talk will bring you a brief overview of the highlights and summaries from #EARLCONF 2024. Wanted to come but didn't make it? Look no further! The talk will feature Hadley Wickham, Author of the Tidyverse's insights on 'R in Production'. Takeaways from our workshops, plus, bonus tips from companies such as Vodafone, Natural England, Amazon, Microsoft and more.

------ General Information ------ Now, refreshments! Drinks and pizza will be served from 6pm. Talks will commence at 6:30pm/6:45pm and the event will conclude at 8:30pm.

The conversation will continue after the event at a local pub TBC – obviously!

Please feel free to register guests and spread the word. We look forward to seeing you all again!

Nous sommes ravis de vous convier à notre prochain événement. Nous aurons le plaisir d’accueillir en personne María Paula Caldas et Julie Aubert, le 29 avril 2024 à partir de 19h dans les locaux de datacraft (https://datacraft.paris/) situés au 3 Rue Rossini, 75009 Paris. Vous pouvez trouver l'adresse sur Google Maps à cette adresse : https://maps.app.goo.gl/1VKXWyhF157xr5Qx6.

Le programme de l’événement est le suivant :

19h - 19h05 : Introduction de la communauté R-Ladies Paris 19h05 - 19h35 : « Création des packages dans R », María Paula Caldas 19h35 - 20h05 : « Développement pratique des modèles d’apprentissage statistique avec {Tidymodels} », Julie Aubert 20h05 - 20h10 : Mot de clôture 20h10 - 20h30 : Réception et échange entre la communauté avec distribution de goodies R

Vous trouverez plus de détails sur nos intervenantes et leurs présentations ci-dessous.

Biographie de la première intervenante : María Paula Caldas est une data scientist à l'OCDE, où elle participe à des projets d'analyse des politiques publiques, de la recherche en économie, et à la production de statistiques internationales. Utilisatrice de R depuis presque 10 ans, elle s'intéresse dernièrement à l'analyse spatiale, au traitement des données avec Arrow, et à la création d'écosystèmes de paquets R pour faciliter le travail en équipe. Investie depuis plusieurs années dans la communauté R, María Paula co-anime actuellement le réseau d'utilisateurs de R/Python/Git de l'OCDE, et contribue (quand elle peut !) à des projets communautaires et open source tels que rOpenSci, R-Ladies, r4ds-es et RencontresR. Description de l’intervention : Suis-je prête à créer mon propre paquet ? Quel est l'intérêt si je travaille en solo ? Comment puis-je partager mon paquet avec les autres ? Nous essaierons d'aborder ces questions ensemble, en partageant nos expériences pendant que nous créons, documentons, testons et partageons un paquet en moins de 30 minutes. Tous les niveaux sont les bienvenus !

Biographie de la deuxième intervenante : Julie Aubert est ingénieure de recherche INRAE en statistiques dans l'équipe SOLsTIS de l’unité MIA Paris-Saclay. Elle travaille dans le domaine de la modélisation et l’apprentissage statistique appliqués aux sciences du vivant avec une spécialisation pour les données omiques en écologie microbienne. Dans ce contexte, elle collabore avec des chercheurs issus de domaines aussi variés que l’agro-écologie, la génétique des plantes ou des bovins, la gestion des risques en agriculture ou l'alimentation et développe des packages R pour diffuser les méthodes proposées. Elle anime également le réseau StateOftheR du département MathNum (https://stateofther.netlify.app/).

Description de l’intervention : Tidymodels regroupe un ensemble de packages R facilitant l’utilisation de méthodes d’apprentissage statistique (telles que les forêts aléatoires, les modèles linéaires ...) dans un cadre unifié et “tidy”. Dans cette présentation, nous verrons comment utiliser ces packages pour prétraiter les données, construire, entraîner et évaluer un modèle, optimiser des hyperparamètres et tout ce qu'il est important de savoir pour mener de bout en bout un projet d’apprentissage statistique supervisé.

Si vous prévoyez de participer à notre événement en distanciel, veuillez vous inscrire via ce formulaire afin de recevoir le lien Zoom 🔗 https://us06web.zoom.us/meeting/register/tZUrdOuuqD0qG9Vx_WPFSj_5VWVxCmbX9dqm.

Cet événement est ouvert à toutes et à tous ! Assurez-vous de réserver votre place dès aujourd'hui.

Nous avons hâte de vous voir le 29 avril à Paris ! 🤩

Please join us July 13th for a virtual seminar on '3 Reasons to use Tidymodels' by Julia Silge!

Modeling and machine learning in R involves a bewildering array of heterogeneous packages, and establishing good statistical practice is challenging in any language. The tidymodels collection of packages offers a consistent, flexible framework for your modeling and machine learning work to address these problems. In this talk, we’ll focus on three specific reasons to consider using tidymodels. We will start with model characteristics themselves, move to the wise management of your data budget, and finish with feature engineering.

All members who RSVP and attend the event will be entered into a raffle and 3 winners will be chosen to receive a free eBook or hard copy of Tidy Modeling with R by Julia Silge and Max Kuuhn (2022)

Please join us July 13th for a virtual seminar on '3 Reasons to use Tidymodels' by Julia Silge! Modeling and machine learning in R involves a bewildering array of heterogeneous packages, and establishing good statistical practice is challenging in any language. The tidymodels collection of packages offers a consistent, flexible framework for your modeling and machine learning work to address these problems. In this talk, we’ll focus on three specific reasons to consider using tidymodels. We will start with model characteristics themselves, move to the wise management of your data budget, and finish with feature engineering.

All members who RSVP and attend the event will be entered into a raffle and 3 winners will be chosen to receive a free eBook or hard copy of Tidy Modeling with R by Julia Silge and Max Kuuhn (2022)

Zoom link TBA.

Get going with tidymodels, a collection of R packages for modeling and machine learning. Whether you're just starting out or have years of experience with modeling, this practical introduction shows data analysts, business analysts, and data scientists how the tidymodels framework offers a consistent, flexible approach for your work. RStudio engineers Max Kuhn and Julia Silge demonstrate ways to create models by focusing on an R dialect called the tidyverse. Software that adopts tidyverse principles shares both a high-level design philosophy and low-level grammar and data structures, so learning one piece of the ecosystem makes it easier to learn the next. You'll understand why the tidymodels framework has been built to be used by a broad range of people. With this book, you will: Learn the steps necessary to build a model from beginning to end Understand how to use different modeling and feature engineering approaches fluently Examine the options for avoiding common pitfalls of modeling, such as overfitting Learn practical methods to prepare your data for modeling Tune models for optimal performance Use good statistical practices to compare, evaluate, and choose among models