Thomas Bury

Building ML and AI systems with Python, LLMs & MLOps.

[Thomas Bury @ThomasJBury]

@steipete @davemorin Typing that on a MacBook?

[Thomas Bury @ThomasJBury]

Promising, converting classical ML models into C native and over efficient APIs 🚀
#machinelearning #python

https://t.co/dbYV7DBMuD

[Thomas Bury @ThomasJBury]

Data should speak for itself, not for the color map. Using the wrong gradient can lead to "optical illusions" where your data looks broken or banded when it is actually smooth.

[Thomas Bury @ThomasJBury]

Scicomap, a Python package for assessing and improving scientific colormaps.

I added a CLI, LLM friendly doc (.md) an interactive @marimo_io tutorial.

Marimo https://t.co/C3q0iAh6Bl
Repo: https://t.co/JJe17PgbrY

#DataVisualization #ScientificComputing #Python #Matplotlib

[Thomas Bury @ThomasJBury]

@RLanceMartin If anyone interestedin the HF implementation: https://t.co/A9RMkzz453

[Colaboratory @GoogleColab]

🚀 Big AI updates in @GoogleColab! We're making generative AI more accessible and easier to use with two new features:
1️⃣ The https://t.co/2QV2vEwI55 Python library is now available to ALL users! Free tier users now get access to Gemini flash models at no cost.
2️⃣ All users get access to a new AI Prompt Cell for easy, low-code AI interaction. Find it in the dropdown next to the "+ Code" button.
See blog below 👇

[Thomas Bury @ThomasJBury]

Interactive Python dashboards—no servers, no hosting fees.

• Author in @marimo_io
• Export as HTML (runs entirely in your browser)
• Push to GitHub—auto-deploys via GitHub Pages and Actions
#Python #DataVisualization #Serverless #GitHubPages
https://t.co/qhdyuqcvPB

[Thomas Bury @ThomasJBury]

@theprasad_ True that

https://t.co/7KKoejMtny

[Thomas Bury @ThomasJBury]

Ever wondered how AI Agents execute code?
https://t.co/A9RMkzz453

#GenAI #AIAgents

[Thomas Bury @ThomasJBury]

@karpathy Using https://t.co/wuqmmV4SwU might help, for having a up to date knowledge base.

Augment code might be better than cursor for large code base

[Thomas Bury @ThomasJBury]

@astral_sh UV is amazing

Here is the part 3, for more advanced usage

https://t.co/LhIs1zqokA

#mlops #python #MachineLearning #SoftwareEngineering #DataScience

[Thomas Bury @ThomasJBury]

Companies on linkedin

[Thomas Bury @ThomasJBury]

UV: Your go-to for Python package & project management. Check out Part 2! #Python #uvpython #pythonpackaging #pyproject #dependencymanagement #pythondev #programming #softwaredevelopment
@astral_sh

https://t.co/BEVc93LzT1

[Thomas Bury @ThomasJBury]

@howdataworks My company (and myself) is using it for feature selection. I guess others used it, it got 100 ⭐

[Thomas Bury @ThomasJBury]

🛠�� Introducing #ARFS: an all-relevant feature selection library designed to find all predictors related to your response—model-agnostic & sklearn compatible! 🚀 Optimize your predictive power. Details 👇
https://t.co/i9pfKNtQCh
#MachineLearning #DataScience #Python #OpenSource

[Thomas Bury @ThomasJBury]

It's time to ditch Poetry (again) 👨‍💻
Thanks to @astral_sh

https://t.co/YKHgqKRmNq

#SoftwareDevelopment #Python

[Rowan Cheung @rowancheung]

Answering community prompts today with Gemini 1.5 Pro.

Comment on the Tweet linked here and I'll try to get your prompt in: https://t.co/o9s2ySc9Wq

[Thomas Bury @ThomasJBury]

The long awaited part 2 on model interpretability is here! #MachineLearning #interpretation #explainableAI #XAI

Interpretable ML 2: From Linear Idols to Non-Linear Frontiers: A Quest for Interpretability https://t.co/QUKbdeS2D2

[Thomas Bury @ThomasJBury]

Generalization of correlation coefficient, applied to feature selection #MachineLearning
#DataScience

Unleash the Potential of Data with ARFS — Part 1: Uncover Hidden Relationships 🔍🚀 https://t.co/3NbL7eXIKU

[Thomas Bury @ThomasJBury]

I've added web-doc for Scicomap! It provides a solution for generating uniform color maps, ensuring your visual information is accurate and appealing. Check it out here: https://t.co/4DezYb6QYS and the companion blog post at https://t.co/RDTujlqKEf #visualization #matplotlib