Lucas de Lima Nogueira

AI

How Bend Works: A Parallel Programming Language That “Feels Like Python but Scales Like CUDA”

A brief introduction to Lambda Calculus, Interaction Combinators, and how they are used to parallelize operations on Bend / HVM. Lucas de Lima Nogueira · Follow Published in Towards Data Science · 22 min read · 10 hours ago — Image by author Introduction If you are reading this article, you probably recently heard about Bend, a new programming language that aims to be massively parallel but without you worrying about things like threads creation,

Read More »
AI

Recreating PyTorch from scratch (with GPU support and automatic differentiation)

Build your own deep learning framework based on C/C++, CUDA, and Python, with GPU support and automatic differentiation Lucas de Lima Nogueira · Follow Published in Towards Data Science · 24 min read · 10 hours ago — Image by Author with the assistance of AI (https://copilot.microsoft.com/images/create) Introduction For many years I have been using PyTorch to construct and train deep learning models. Even though I have learned its syntax and rules, something has always

Read More »
AI

Why Deep Learning Models Run Faster on GPUs: A Brief Introduction to CUDA Programming

For those who want to understand what .to(“cuda”) does Lucas de Lima Nogueira · Follow Published in Towards Data Science · 15 min read · Apr 17, 2024 — Image by the author with the assistance of AI (https://copilot.microsoft.com/images/create) Nowadays, when we talk about deep learning, it is very common to associate its implementation with utilizing GPUs in order to improve performance. GPUs (Graphical Processing Units) were originally designed to accelerate rendering of images, 2D,

Read More »