sayzardFeb 12

möchtegern (@heft_ig)

'Neural Networks: Zero to Hero'라는 교육 자료(강의 맵)가 Karpathy의 강의/비디오와 직접 매핑된다는 안내 트윗입니다. 문서(또는 페이지)는 micrograd(학습용 autograd 구현) 설명과 microgpt.py 예제, 그리고 Claude Opus 4.6 언급을 포함해 신입·개발자 대상 학습 경로를 제시합니다.

https://x.com/heft_ig/status/2021798285992763430

#micrograd #microgpt #claudeopus #karpathy #neuralnetworks

möchtegern (@heft_ig) on X

@karpathy "Neural Networks: Zero to Hero" Every section maps directly to one or more of Karpathy's lectures/videos Here's the breakdown(Claude opus 4.6) https://t.co/GyoLTrXsim 1. **micrograd** (https://t.co/7LFTYUqZWw) -- to understand the `Value` class and `backward()` (the engine that

X (formerly Twitter)
sayzardFeb 11

Andrej Karpathy (@karpathy)

전체 LLM 아키텍처와 손실 함수를 가장 원자적인 수학 연산(+ ,*,**, log, exp) 수준으로 완전히 분해하고, 작은 스칼라값 autograd 엔진(micrograd)으로 그래디언트를 계산하며 Adam 옵티마이저를 사용하는 구현/교육적 접근을 설명하는 기술적 트윗입니다.

https://x.com/karpathy/status/2021695367507529825

#autograd #micrograd #llm #optimization

Andrej Karpathy (@karpathy) on X

The way it works is that the full LLM architecture and loss function is stripped entirely to the most atomic individual mathematical operations that make it up (+, *, **, log, exp), and then a tiny scalar-valued autograd engine (micrograd) calculates gradients. Adam for optim.

X (formerly Twitter)
Show threadJuliuFeb 18, 2023

Micrograd is very simple, only fully connected layers. So first trying to find out if it can even learn numbers based on MNIST dataset.

Then I hope to at least be able to verfit, so the essence works. Then I'll have the challenge of trying to make it work for every icon in the app.

Presumably I have to create/generate a huge set of icon images to train on..

#ux #micrograd #reactNative #MLP #MNIST #TinyUX #Karpathy #ai #NeuralNet