When do you have "enough data" to make a confident decision? 📊
@mena_wang dives into how statistical power and sample size help you confidently separate signal from noise, even when data is messy.
https://t.co/T7XHbBW1qR
Did you know that very different business realities can hide behind the same bar chart — one of the most common ways insights are presented in the business world?
@mena_wang shares practical steps to analyze group differences and avoid costly misinterpretations.
https://t.co/GrYDWy6TIE
Struggling with model performance? Learn how to identify and manage the problematic data points, not just your data. @mena_wang unveils a Cook's Distance-inspired approach to pinpoint these influential observations.
https://t.co/66s6Gg9zfK
That 5-point difference in your dashboard might be meaningless. @mena_wang's new article reveals how a simple bar chart can hide three distinct business realities and lead to misinterpretations.
https://t.co/T7XHbBW1qR
Build more robust and reliable ML models. @mena_wang's guide to identifying influential observations teaches you to move beyond simple outlier detection, equipping you with a diagnostic tool for your workflow.
https://t.co/66s6Gg9zfK
With this helper function, you can easily detect and diagnose the most influential points in the training set that pull the model’s predictions away from the “general pattern” reflected in the rest of the data.
Learn more on TDS: https://t.co/tokDq9brzS
Worried your model is being held hostage by a few bad apples? Learn how to identify and neutralize influential data points that cripple generalization. @mena_wang unveils a smarter approach to anomaly detection.
https://t.co/66s6Gg9zfK
Learn more about MLarena 📦, an open source Python package for streamlining your ML workflow in this article published on @TDataScience
https://t.co/QXLMtISnvm
🧵4/4
Would you like to have a comprehensive evaluation report built for your ML model in just one line? 😎
This is what the `evaluation` method in `MLarena` produces for any sklean-style classification estimators. 😍
🧵 1/4
#MachineLearning#DataScience#pythonlearning#MLOps
Tired of ML boilerplate? @mena_wang's new article introduces MLarena, an open-source Python package that builds algorithm-agnostic ML pipelines in a breeze! Discover how it streamlines training, diagnostics, and optimization.
https://t.co/vHpu031cVV
Just created a site to offer #Python to #R translation of 30 essential #Pandas methods in The Only 30 Methods You Should Master To Become A Pandas Pro published on @TDataScience
Site and source code here if interested: https://t.co/jpFGypzu5b
#rstats#DataScience#tidyverse
Introducing an upgraded Claude 3.5 Sonnet, and a new model, Claude 3.5 Haiku. We’re also introducing a new capability in beta: computer use.
Developers can now direct Claude to use computers the way people do—by looking at a screen, moving a cursor, clicking, and typing text.
Looking for a continuing position as Lecturer in Statistics? Come join us @ANU_CBE! Job application closes on July 30th Canberra time. 🎓
https://t.co/7w6SO5PwGB
Consider being a volunteer puppy carer?
Me and my family have cared for three SED puppies till now. They have brought us so much joy! ❤️And it warms our hearts to think what an important role they may play in another person's life. 🐕🦺
More info here: https://t.co/ifupZodQZT
Vision Australia is on a mission - to urgently find homes for 45 puppies who'll grow up to be Seeing Eye Dogs.
Mother Nature has delivered a puppy boom resulting in a record breeding year, but there's a shortage of volunteer carers. @Jo_Hall9#9News