@towardsdatascience@me.dm
@towardsdatascience@me.dm avatar

towardsdatascience

@towardsdatascience@me.dm

A Medium publication sharing concepts, ideas, and codes. Share your insights and projects with our global audience: http://bit.ly/write-for-tds.

This profile is from a federated server and may be incomplete. Browse more on the original instance.

towardsdatascience, to random
@towardsdatascience@me.dm avatar

After unpacking the inner workings of the Segment-Anything model's decoder, Wei Yi is back with another comprehensive explainer, this time focusing on the model's encoder. https://buff.ly/3wCul10

towardsdatascience, to random
@towardsdatascience@me.dm avatar

"Do you know it’s possible to fine-tune a Language Model using just a few parameters and a tiny dataset with as few as 10 data points?

Well, it’s not magic."

🖋️ by Yanli Liu https://buff.ly/3wRrIsg

towardsdatascience, to random
@towardsdatascience@me.dm avatar

In his latest model-focused explainer, W Brett Kennedy zooms in on additive decision trees, an often better-performing variant of traditional decision trees. https://buff.ly/3WX83Su

towardsdatascience, to random
@towardsdatascience@me.dm avatar

Interested in learning how you can optimize operational queues with the help of simulations? Don't miss Mariya Mansurova's latest deep dive geared towards data-focused product analysts. https://buff.ly/4brZPGe

towardsdatascience, to random
@towardsdatascience@me.dm avatar

How can data scientists navigate the specific types of stress their role often involves?

Zijing Zhu shares a thoughtful reflection on stress management, keeping things in perspective, and adopting habits that can help you maintain a sustainable rhythm at work. https://buff.ly/3QXF10Y

towardsdatascience, to random
@towardsdatascience@me.dm avatar

In the mood for some hands-on tinkering?

Don't miss our best recent project walkthroughs, with top-notch contributions by Anand Subramanian, Claudia Ng, Maria Mouschoutzi, Tahreem Rasul, Alok Suresh, Theophano Mitsa, Yuan Tian, Deepsha Menghani, and Lucas de Lima Nogueira. https://buff.ly/4dWZ49G

towardsdatascience, to random
@towardsdatascience@me.dm avatar

Looking for a hearty, thought-provoking article to dig into this weekend (and beyond)? Don't miss Dusko Pavlovic's deep dive on the core theoretical and mathematical concepts behind the idea of learning (whether by humans or machines). https://buff.ly/3VdciYI

towardsdatascience, to random
@towardsdatascience@me.dm avatar

How should you go about detecting outliers in your time-series data? Sara Nóbrega's comprehensive overview presents effective statistical methods and tools for you to integrate into your workflows. https://buff.ly/3yHj8g7

towardsdatascience, to random
@towardsdatascience@me.dm avatar

What can data engineers do to support reduced carbon emissions as a result of their work?

Hussein Jundi shares a detailed framework for adopting sustainable patterns and translating them into real-world workflows. https://buff.ly/44FGkr2

towardsdatascience, to random
@towardsdatascience@me.dm avatar

Musical data is much tougher to come by than textual and visual data. Max Hilsdorf's practical guide is here to help, with three actionable approaches for overcoming data scarcity for your AI tool or project. https://buff.ly/3VciSP6

towardsdatascience, to random
@towardsdatascience@me.dm avatar

Learn How to Enhance Your Data Analysis for Advanced Computational Tasks, from Innovative Optimization Strategies to Foundational Machine Learning Algorithms.

🖋️ by Sydney Nye https://buff.ly/3yomiFy

towardsdatascience, to random
@towardsdatascience@me.dm avatar

In a detailed and patient guide, Anand Subramanian outlines a practical approach for building automated clinical coding systems with LLMs. https://buff.ly/3ylNzIK

towardsdatascience, to random
@towardsdatascience@me.dm avatar

Large Language Models have their strengths, but for many production problems, simpler NLP techniques are faster, cheaper, and just as effective.

🖋️ by Katherine Munro https://towardsdatascience.com/yes-you-still-need-old-school-nlp-skills-in-the-age-of-chatgpt-a26a47dc23d7

towardsdatascience, to random
@towardsdatascience@me.dm avatar

In this article, Marcin Kozak explains how to improve your Python code with simple, readable and performant enumerations: https://towardsdatascience.com/a-guide-to-powerful-python-enumerations-f72f185b6883

towardsdatascience, to random
@towardsdatascience@me.dm avatar

"We will see that, unlike other XAI methods like SHAP, LIME, ICE Plots and Friedman’s H-stat, ALEs give interpretations that are robust to multicollinearity."

Deep Dive on Accumulated Local Effect Plots (ALEs) with Python by Conor O'Sullivan https://towardsdatascience.com/deep-dive-on-accumulated-local-effect-plots-ales-with-python-0fc9698ed0ee

towardsdatascience, to random
@towardsdatascience@me.dm avatar

In this tutorial, Katia Gil Guzman shows how to transform a relational database into a dynamic graph database in Python: https://towardsdatascience.com/turning-your-relational-database-into-a-graph-database-c4cee3d5c6d2

towardsdatascience, to random
@towardsdatascience@me.dm avatar

What metrics should you choose to evaluate the performance of your model's predictions? Jeffrey Näf provides a nuanced guide to a crucial step in any ML workflow. https://buff.ly/3yxrtTv

towardsdatascience, to random
@towardsdatascience@me.dm avatar

"In this article, I’ll describe in more detail how this physics-centric understanding can help us gain more insight into our data."

The Physics Behind Data by Tim Lou, PhD https://buff.ly/3K5TnIT

towardsdatascience, to random
@towardsdatascience@me.dm avatar

What is the most effective approach for encoding hierarchical categoricals? Valerie Carey's new deep dive tests alternatives to hierarchical blending in an XGBoost model. https://buff.ly/44Ol7ex

towardsdatascience, to random
@towardsdatascience@me.dm avatar

Our weekly highlights are all about the new frontiers of LLMs, from open-source foundation models to temperature scaling. Don't miss top-notch contributions by Leonie Monigatti, Parul Pandey, Mike Cvet, Leon Eversberg, and Eyal Aharoni & Eddy Nahmias. https://buff.ly/3K48gf1

towardsdatascience, to random
@towardsdatascience@me.dm avatar

Wouldn't it be neat if you could detect and fix any type of error in a directed acyclic graph (DAG) to ensure that it accurately represents the underlying data? Graham Harrison dug deep into the research and presents a promising "unified theory of everything" in the context of causal inference. https://buff.ly/452bqcV

towardsdatascience, to random
@towardsdatascience@me.dm avatar

"We will tackle the task of simulating processes to intelligently manage the coming and going of birds through object detection and water sprinkling subsystems."

Sofya Lipnitskaya continues her excellent series on finite-state machine modeling for real-world AI systems. https://buff.ly/3wNZnmx

towardsdatascience, to random
@towardsdatascience@me.dm avatar

Learn what constitutes a multi-arm bandit problem and how to go about solving one effectively — Jarom Hulet's new deep dive offers a fishing-inspired example to help you intuitively grasp the earn/learn trade-off. https://buff.ly/44JRN9a

towardsdatascience, to random
@towardsdatascience@me.dm avatar

The story of a 19th-century shipwreck forms the basis of Sachin Date's excellent new deep dive on statistical convergence, which covers its mathematical underpinnings in great detail. https://buff.ly/4bIU2vT

towardsdatascience, to random
@towardsdatascience@me.dm avatar

"With interpretable models, we do not have these issues. The model is itself comprehensible and we can know exactly why it makes each prediction."

W Brett Kennedy introduces ikNN — interpretable k Nearest Neighbors — as a new alternative to available models. https://buff.ly/3QIQ26c

  • All
  • Subscribed
  • Moderated
  • Favorites
  • •
  • megavids
  • Durango
  • DreamBathrooms
  • ngwrru68w68
  • osvaldo12
  • cubers
  • magazineikmin
  • ethstaker
  • Youngstown
  • rosin
  • slotface
  • everett
  • kavyap
  • InstantRegret
  • JUstTest
  • khanakhh
  • thenastyranch
  • anitta
  • modclub
  • GTA5RPClips
  • mdbf
  • cisconetworking
  • tester
  • tacticalgear
  • provamag3
  • Leos
  • normalnudes
  • lostlight
  • All magazines
    •