Emmanuel Ikpesu

Nobody knows the future, and where opportunity waits. But uncertainty is not a reason not to take action. So be prepared for when an opportunity comes. Rooting for you🫶

I’ve built many interactive #Python dashboards with @matplotlib to teach spatial continuity quantification through variogram calculation and modeling 📊🐍Here’s one example where I demonstrate how to model zonal anisotropy, a commonly encountered spatial structure in the presence of trends and bedding ⛏️🪨 🌍I’m stoked!! Spatial continuity quantification is so powerful—and once you can model it, you can make way better decisions and deliver way more impact. 🚀 🔥Totally stoked to share this! 😄✨

@Fowe Hi I’m interested. I try reaching out but your DM is locked.

If you know strong graduate students in the USA still looking for Data science or AI/ML engineering internship job for this summer please DM me. Please recommend only if you can vouch for the candidate. Thanks

Did you make a #NewYearsResolution to learn #DataScience? I’m here to help guide your 2026 learning journey 🚀 I openly share all of my university-level educational content with the world—carefully curated courses that include: • YouTube lectures, organized into playlists • Free, online course e-books • Python demonstration workflows and interactive dashboards All designed with one goal in mind, maximize your learning success. Get started at: 1. Data Analytics and Geostatistics - geostatsguy.github.io/GeostatsPyDemo… 2. Machine Learning - geostatsguy.github.io/MachineLearnin…

Today at 5 PM (ETC 2.136) on the @UTAustin campus, we’re hosting our first information session for the Energy AI 2026 Hackathon, co-hosted by @GeostatsGuy & @johntfoster. The event is open to all current UT Austin students—both undergraduate and graduate! This will be our 6th annual hackathon, which typically brings together 150+ Longhorns from across campus, 20+ industry mentors, and 6 executive judges from industry. The event is proudly supported by @UT_PGE Chair Matt Balhoff and our amazing staff. This year, we’re excited to welcome our hackathon architects, Dr. Nataly Chacon Buitrago and Dinghan Wang! Are you a current #Longhorn interested in learning and applying #DataScience to a challenging spatiotemporal, multivariate problem? Come by today to learn more! 🧠⚡️

I’m stoked! My #DataAnalytics and #Geostatistics course for Spring 2026 is already filling up with students excited to dive into #spatial #data and its real-world impact. Yes, the classroom is almost full, but the best part is this, I make all of the content freely available to anyone in #Texas and beyond. The world’s biggest challenges are spatial, and geostatistics (theory and practice) is essential for the prosperity, safety, and health of communities everywhere. I’m grateful for the chance to teach it and to share these tools as broadly as possible. Free online, e-book: geostatsguy.github.io/GeostatsPyDemo… Course Lectures: youtube.com/playlist?list=… #WhatStartsHere

-Help ya girl repost? 🥺❤️- Hey yall, if you’re looking for some custom-made, pinterest-inspired, cutesy, aesthetic stuff like keychains or bracelets you can check out my IG profile or DM me for more info 🫶🏼 Morse code bracelets : RM10 Custom beaded clay keychain : RM15

Season 3. Are you ready? 🔥🔥

Howdy folks! I’m Michael Pyrcz, a proud Ukrainian-Canadian professor thrilled to be here in Texas teaching data science to students across our campus. To me, being a professor is first and foremost a role of service. That’s why I’m dedicated to sharing my educational content freely with anyone eager to learn. Whether it’s my lectures on YouTube, reproducible course materials on GitHub, or my two free, online e-books—which I’m continuously updating—you’re welcome to explore and learn at your own pace! Check out the books here: Data Analytics and Geostatistics: geostatsguy.github.io/GeostatsPyDemo… Machine Learning: geostatsguy.github.io/MachineLearnin…

@Faith_Tife03 @iam_Uchenna That’s awesome! I’m not a fan of R2 because it’s valid when you assume linear relationship. And NN are parametric, so they don’t make that linearity assumption except you are using all linear activation function which would eventually collapse your model to a linear regression.

@TheDataHokage @iam_Uchenna Thank you. I used mae, mse, rmse and r2 for the evaluation as I would usually use for regression. The model is actually doing well at the moment, just looking out to optimize a bit.

what are you working on?

@Faith_Tife03 @iam_Uchenna For DL, RMSE might be better than R2, expect you’re certain you have a linear relationship. If you have a linear relationship, just use linear regression except you’re trying to learn DL. In all, visualize your results to see how your model is performing.

@iam_Uchenna For DL, R^2 is 0.86 . I need to do better. For the Bayesian, model is doing well. I am battling with the visualization of the prediction.(Bayesian Network)

@Fortunekonnect 🤣 Superman return!

@TheDataHokage Ikpesu is back on X ke?

Deep learning is linear regression with activation function.🤧

@don_se7en Remove the activation function, and it would collapse to linear regression. Fun fact: for simple models, linear regression is better.

Yeah tbh, many Linear regression models feeding into each other and activation functions to add non-linearity.

Welcome to Mars

Using LLM to train LLM is wild!

🚀 I'm hiring an Executive Assistant 🚀 Looking for a driven individual to work directly with me at Paystack, Sporting Lagos, and my investments. 📅 12-18 month paid fellowship 📍 Location: Lagos, Nigeria 💪 Role: Project mgmt, travel/logistics, & personal office ops If you have great taste, empathy, & thrive in ambiguity, apply below! 👇

To help my students comprehend artificial neural networks (ANN), including initialization, backpropagation, and updating, I built an ANN from scratch and developed a custom interactive #Python dashboard using @matplotlib. Now, when I teach these concepts, students engage in hands-on, experiential learning by interacting directly with the machine. It’s a fun and effective way to make complex topics more accessible! I share it on #GitHub @ github.com/GeostatsGuy/Da… ∀. #DataScience #MachineLearning

During the American Thanksgiving break, I had some valuable time to build out the "#Probability Concepts" chapters for my "Applied #Geostatistics" and "Applied #MachineLearning" free, e-books. It's been an exciting and productive experience! I'm putting in a lot of effort to ensure that the essential prerequisites are clear and accessible to everyone. I'll be updating the e-books this coming Monday. I'm stoked to share evergreen educational content!

🚀 I am happy to announce that I am rejoining @Microsoft as a Senior Product Marketing Manager, Power Platform in Redmond. With Copilot in Power Platform, I am happy to drive awareness of AI capabilities in Power Platform stack through community engagement

FULL-TIME | A point apiece. 🩵 3-3 ⚫️ #ManCity | #UCL