Star200+

์Šคํ‚ฌ ๊ฐ์‚ฌ ๋ณด๊ณ ์„œ

Annotated Deep Learning Paper Implementations ๊ฐ์‚ฌ ๋ณด๊ณ ์„œ.

๐Ÿง‘โ€๐Ÿซ 60+ Implementations/tutorials of deep learning papers with side-by-side notes ๐Ÿ“; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, sophia, ...), gans(cyclegan, stylegan2, ...), ๐ŸŽฎ reinforcement learning (ppo, dqn), capsnet, distillation, ... ๐Ÿง 

๊ฒ€ํ† ๋จ ยท ๊ฒ€ํ† ๊ฒ€ํ†  ํ•„์š”์ƒ์„ฑ๋จ 2026๋…„ 8์›” 22์ผํœด๋ฆฌ์Šคํ‹ฑ ๋ฉ”ํƒ€๋ฐ์ดํ„ฐ ๊ฐ์‚ฌ
89
๊ฐ์‚ฌ
89
์‹ ๋ขฐ
100
ํ’ˆ์งˆ
89
๋ณด์•ˆ
62
์œ ์ง€๋ณด์ˆ˜
92
์„ค์น˜

OpenAgentSkill ์‹ ๋ขฐ ์ ์ˆ˜

89
ํ”„๋กœ๋•์…˜ ํ›„๋ณด

OpenAgentSkill ์‹ ๋ขฐ ์ ์ˆ˜

Trust Score๋Š” ์„ค์น˜ ์ „์— ํ›„๋ณด ๋ชฉ๋ก์— ๋„ฃ์„ ๋งŒํผ ์•ˆ์ „ํ•œ์ง€ Agent๊ฐ€ ํŒ๋‹จํ•˜๋„๋ก ๋•์Šต๋‹ˆ๋‹ค.

GitHub ์ฑ„ํƒ๋„

ํ†ต๊ณผ

100

GitHub ์Šคํƒ€ 67K

์Šคํƒ€/ํฌํฌ ํ™œ๋™

ํ†ต๊ณผ

100

์Šคํƒ€ 67K, ํฌํฌ 6.7K; ํ˜„์žฌ ๋ฉ”ํƒ€๋ฐ์ดํ„ฐ์—์„œ ์ด์Šˆ ํ™œ๋™์„ ํ™•์ธํ•  ์ˆ˜ ์—†์Šต๋‹ˆ๋‹ค

์ตœ๊ทผ ์œ ์ง€๋ณด์ˆ˜

์ •๋ณด

62

๋งˆ์ง€๋ง‰ ํ‘ธ์‹œ ํ›„ 7๊ฐœ์›”

๋ผ์ด์„ ์Šค ๋ช…ํ™•์„ฑ

ํ†ต๊ณผ

86

MIT

README/SKILL.md ์™„์„ฑ๋„

ํ†ต๊ณผ

90

๋ฉ”ํƒ€๋ฐ์ดํ„ฐ์— ์ถฉ๋ถ„ํ•œ ์‚ฌ์šฉ ๋ฐ ์›Œํฌํ”Œ๋กœ ๋งฅ๋ฝ์ด ํฌํ•จ๋˜์–ด ์žˆ์Šต๋‹ˆ๋‹ค

์˜์กด์„ฑ/๋Ÿฐํƒ€์ž„ ์œ„ํ—˜

ํ†ต๊ณผ

90

๊ณต๊ฐœ ๋ฉ”ํƒ€๋ฐ์ดํ„ฐ์— ์ฃผ์š” ์˜์กด์„ฑ ์œ„ํ—˜ ํžŒํŠธ๊ฐ€ ์—†์Šต๋‹ˆ๋‹ค

์„ค์น˜ ๊ฐ€๋Šฅ ์—ฌ๋ถ€

ํ†ต๊ณผ

92

npx skills add labmlai/annotated_deep_learning_paper_implementations

์„ค์น˜ ๋ช…๋ น ์•ˆ์ „์„ฑ

ํ†ต๊ณผ

92

ํ‘œ์ค€ ํŒจํ‚ค์ง€ ๋˜๋Š” ๋Ÿฐํƒ€์ž„ ์„ค์น˜ ๊ฒฝ๋กœ

๊ถŒํ•œ ๋ฒ”์œ„

ํ†ต๊ณผ

86

ํŒŒ์ผ ์‹œ์Šคํ…œ ๋˜๋Š” ๋ฌธ์„œ ์ ‘๊ทผ

์ €์žฅ์†Œ ๊ทผ๊ฑฐ

ํ†ต๊ณผ

86

https://github.com/labmlai/annotated_deep_learning_paper_implementations

๊ฒ€ํ†  ์ƒํƒœ

ํ†ต๊ณผ

88

AI ๊ฒ€ํ†  ๋ฐ์ดํ„ฐ๋ฅผ ์‚ฌ์šฉํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค

Agent ๊ฒ€์ฆ ๊ฒฐ๊ณผ

์ •๋ณด

54

์•„์ง Agent ๊ฒฐ๊ณผ ๋ฐ์ดํ„ฐ๊ฐ€ ์—†์Šต๋‹ˆ๋‹ค

๊ฒ€์‚ฌ

์„ค์น˜ ๋ฐ ์ฑ„ํƒ ๊ฒ€ํ† 

10 ํ†ต๊ณผ ยท 4 ๊ฒ€ํ†  ํ•„์š”

์„ค์น˜ ๊ฒฝ๋กœ

92

ํ†ต๊ณผ

npx skills add labmlai/annotated_deep_learning_paper_implementations

์ €์žฅ์†Œ

88

ํ†ต๊ณผ

https://github.com/labmlai/annotated_deep_learning_paper_implementations

๋ผ์ด์„ ์Šค

86

ํ†ต๊ณผ

MIT

์œ ์ง€๋ณด์ˆ˜

62

ํ™•์ธ

๋งˆ์ง€๋ง‰ ํ‘ธ์‹œ ํ›„ 7๊ฐœ์›”

AI ๊ฒ€ํ† 

88

ํ†ต๊ณผ

Approved with no listed issues

README/SKILL.md ์™„์„ฑ๋„

90

ํ†ต๊ณผ

Usable description available

์˜์กด์„ฑ ์œ„ํ—˜

90

ํ†ต๊ณผ

๊ณต๊ฐœ ๋ฉ”ํƒ€๋ฐ์ดํ„ฐ์— ์ฃผ์š” ์˜์กด์„ฑ ์œ„ํ—˜ ํžŒํŠธ๊ฐ€ ์—†์Šต๋‹ˆ๋‹ค

์„ค์น˜ ๋ช…๋ น ์•ˆ์ „์„ฑ

92

ํ†ต๊ณผ

ํ‘œ์ค€ ํŒจํ‚ค์ง€ ๋˜๋Š” ๋Ÿฐํƒ€์ž„ ์„ค์น˜ ๊ฒฝ๋กœ

๊ถŒํ•œ ๋ฒ”์œ„

86

ํ†ต๊ณผ

ํŒŒ์ผ ์‹œ์Šคํ…œ ๋˜๋Š” ๋ฌธ์„œ ์ ‘๊ทผ

์Šคํƒ€/ํฌํฌ ํ™œ๋™

100

ํ†ต๊ณผ

์Šคํƒ€ 67K, ํฌํฌ 6.7K; ํ˜„์žฌ ๋ฉ”ํƒ€๋ฐ์ดํ„ฐ์—์„œ ์ด์Šˆ ํ™œ๋™์„ ํ™•์ธํ•  ์ˆ˜ ์—†์Šต๋‹ˆ๋‹ค

์ฑ„ํƒ๋„

88

ํ†ต๊ณผ

GitHub ์Šคํƒ€ 67K

Financial decision safety

58

ํ™•์ธ

Research-only use: do not treat output as financial advice or execute a position without human approval.

๊ฒฝ๊ณ 

  • Financial research output is not financial advice; require human review before any live investment decision
  • Financial research output is not financial advice; require human review before any live investment decision.

๋ฐฉ๋ฒ•

์ด ๋ณด๊ณ ์„œ๋Š” ๊ณต๊ฐœ ๋ฉ”ํƒ€๋ฐ์ดํ„ฐ, AI ๊ฒ€ํ†  ๊ฒฐ๊ณผ, ์ €์žฅ์†Œ ์ตœ์‹ ์„ฑ, ์„ค์น˜ ์ค€๋น„ ์ƒํƒœ, OpenAgentSkill ์ด๋ฒคํŠธ, ํ’ˆ์งˆ ์ ์ˆ˜, ์‹ ๋ขฐ ๊ฒ€์‚ฌ์™€ Agent ์•ˆ์ „ ๊ฒŒ์ดํŠธ๋ฅผ ๊ฒฐํ•ฉํ•ฉ๋‹ˆ๋‹ค. ์ „์ฒด ์†Œ์Šค ์ฝ”๋“œ ๋ณด์•ˆ ๊ฐ์‚ฌ๋Š” ์•„๋‹™๋‹ˆ๋‹ค.

๊ฐ€๊นŒ์šด ์˜ต์…˜ ๋น„๊ต

๋‹ค์Œ์œผ๋กœ ๊ฐ์‚ฌํ•  ๊ด€๋ จ ์Šคํ‚ฌ

Tensorflow

An Open Source Machine Learning Framework for Everyone

196K ์Šคํƒ€ ยท ๊ฐ์‚ฌ ๋ณด๊ณ ์„œ

Transformers

๐Ÿค— Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

162K ์Šคํƒ€ ยท ๊ฐ์‚ฌ ๋ณด๊ณ ์„œ

Pytorch

Tensors and Dynamic neural networks in Python with strong GPU acceleration

101K ์Šคํƒ€ ยท ๊ฐ์‚ฌ ๋ณด๊ณ ์„œ