728x90
๋ฐ˜์‘ํ˜•

๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I. 23

[Python][AI] ๋กœ๋˜ ๋ฒˆํ˜ธ ์˜ˆ์ธก ํ…Œ์ŠคํŠธ ์ง„ํ–‰ ๋ฐ ๊ฒฐ๊ณผ ๋ถ„์„

2025.02.18 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] ํ•œ๊ตญ ๋กœ๋˜ ๋ถ„์„: ๋‹น์ฒจ ํ™•๋ฅ ๊ณผ ์˜ˆ์ธก์˜ ๋ถˆ๊ฐ€๋Šฅ์„ฑ2025.02.19 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] ํ•œ๊ตญ ๋กœ๋˜ ๋ถ„์„: ๋‹น์ฒจ ๋ฒˆํ˜ธ ๋ถ„์„๊ณผ ํŒจํ„ด ์ฐพ๊ธฐ(EDA)2025.02.24 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] ํ•œ๊ตญ ๋กœ๋˜ ๋ถ„์„ : ์ถ”๊ฐ€ EDA ๋ฐ ML ๋ฒˆํ˜ธ ์˜ˆ์ธก ๋กœ๋˜ ๋ฒˆํ˜ธ ์˜ˆ์ธก์„ ์œ„ํ•œ ์ถ”๊ฐ€ ํ…Œ์ŠคํŠธ๋ฅผ ์ง„ํ–‰ํ•˜์˜€๋‹ค. ์ด๋ฒˆ ํ…Œ์ŠคํŠธ์—์„œ๋Š” ๋จธ์‹ ๋Ÿฌ๋‹ ๊ธฐ๋ฐ˜ ๋ชจ๋ธ์„ ํ™œ์šฉํ•˜์—ฌ ๋กœ๋˜ ๋‹น์ฒจ ๋ฒˆํ˜ธ๋ฅผ ์˜ˆ์ธกํ•˜๊ณ , ๊ทธ ์„ฑ๋Šฅ์„ ํ‰๊ฐ€ํ•˜๋Š” ๊ณผ์ •์„ ์ง„ํ–‰ํ•˜์˜€๋‹ค.1. ์˜ˆ์ธก ๋ชจ๋ธ ๊ฐœ์š”์˜ˆ์ธก์„ ์œ„ํ•ด CatBoost, XGBoost, LightGBM, RandomForest ์ด 4๊ฐ€์ง€ ๋จธ์‹ ๋Ÿฌ๋‹ ๋ชจ๋ธ์„ ์‚ฌ์šฉํ•˜..

[Python][AI] ์•„๊ธฐ์˜ ์‹ฌ์žฅ ์†Œ๋ฆฌ๋กœ ์„ฑ๋ณ„ ์˜ˆ์ธก? – ๋ฐ์ดํ„ฐ์…‹๊ณผ ๋…ผ๋ฌธ ์†Œ๊ฐœ

์•„๋‚ด๊ฐ€ ์ž„์‹ ํ•˜์—ฌ ์–ผ๋งˆ์ „์— ์‚ฐ๋ถ€์ธ๊ณผ๋ฅผ ๋‹ค๋…€์™”์Šต๋‹ˆ๋‹ค. ์•„์ง 1.5cm์— ๋ถˆ๊ณผํ•œ ํƒœ์•„์˜ ์‹ฌ์žฅ ์†Œ๋ฆฌ๋ฅผ ๋“ค์œผ๋‹ˆ ์‹ฌ์ •์„ ์ด๋ฃจ ๋งํ•˜๊ธฐ ์–ด๋ ต๋”๊ตฐ์š”. ๋ณ‘์›์„ ๋‚˜์˜ค๊ณ  ์นœ๊ตฌ๋“ค๊ณผ ์–˜๊ธฐ๋ฅผ ๋‚˜๋ˆ ๋ณด๋‹ˆ ์‹ฌ์žฅ ์†Œ๋ฆฌ๋ฅผ ๋“ฃ๊ณ  ์•„๋“ค์ธ์ง€ ๋”ธ์ธ์ง€ ํŒ๋‹จํ•  ์ˆ˜ ์žˆ๋‹ค๊ณ  ํ•ฉ๋‹ˆ๋‹ค. ๊ธฐ์ฐจ์†Œ๋ฆฌ์™€ ๋น„์Šทํ•˜๋‹ค๋ฉด ์•„๋“ค, ๋ง๋ฐœ๊ตฝ ์†Œ๋ฆฌ๊ฐ€ ๋‚œ๋‹ค๋ฉด ๋”ธ์ด๋ผ๋Š” ์–˜๊ธฐ๋ฅผ ๋“ค์–ด ๋ณด์…จ๋‚˜์š”? ์ด์— ํ˜ธ๊ธฐ์‹ฌ์ด ์ƒ๊ฒจ ์—ฐ๊ตฌ ๊ฒฐ๊ณผ๊ฐ€ ์žˆ๋Š”์ง€ ์ฐพ์•„๋ณด์•˜๊ณ , ์ฐธ๊ณ ํ• ๋งŒํ•œ ๋ฐ์ดํ„ฐ์…‹๊ณผ ์—ฐ๊ตฌ ๊ฒฐ๊ณผ๋ฅผ ์ฐพ์„ ์ˆ˜ ์žˆ์—ˆ์Šต๋‹ˆ๋‹ค. ๋ฐ์ดํ„ฐ์…‹๊ณผ ์—ฐ๊ตฌ ๊ฒฐ๊ณผ๋ฅผ ์†Œ๊ฐœํ•˜๊ณ  ์‹ค์Šต์„ ์ง„ํ–‰ํ•˜๋ฉฐ ์•„๋“ค์ผ์ง€ ๋”ธ์ผ์ง€ ํŒ๋ณ„ํ•˜๋Š” ์‹ค์Šต์„ ์ง„ํ–‰ํ•˜๊ณ ์ž ํ•ฉ๋‹ˆ๋‹ค.1. Shiraz University Fetal Heart Sounds Database(SUFHSDB) ์†Œ๊ฐœSUFHSDB๋Š” ์ด๋ž€ Shiraz University์—์„œ ์ˆ˜์ง‘ํ•œ ํƒœ์•„ ๋ฐ ์‚ฐ๋ชจ์˜ ์‹ฌ..

[Python][AI] ํ•œ๊ตญ ๋กœ๋˜ ๋ถ„์„ : ์ถ”๊ฐ€ EDA ๋ฐ ML ๋ฒˆํ˜ธ ์˜ˆ์ธก

2025.02.18 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] ํ•œ๊ตญ ๋กœ๋˜ ๋ถ„์„: ๋‹น์ฒจ ํ™•๋ฅ ๊ณผ ์˜ˆ์ธก์˜ ๋ถˆ๊ฐ€๋Šฅ์„ฑ2025.02.19 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] ํ•œ๊ตญ ๋กœ๋˜ ๋ถ„์„: ๋‹น์ฒจ ๋ฒˆํ˜ธ ๋ถ„์„๊ณผ ํŒจํ„ด ์ฐพ๊ธฐ(EDA) ๋กœ๋˜ ๋ฐ์ดํ„ฐ ๋ถ„์„์„ ํ†ตํ•ด ๋‹น์ฒจ ๋ฒˆํ˜ธ์˜ ํŒจํ„ด์„ ์ฐพ์•„๋ณด๊ณ , XGBoost๋ฅผ ํ™œ์šฉํ•˜์—ฌ ๋‹ค์Œ ๋‹น์ฒจ ๋ฒˆํ˜ธ๋ฅผ ์˜ˆ์ธกํ•˜๋Š” ๋ฐฉ๋ฒ•์„ ์†Œ๊ฐœํ•œ๋‹ค. ์ด๋ฒˆ ๋ถ„์„์—์„œ๋Š” ํ™€์ˆ˜/์ง์ˆ˜ ๋น„์œจ, ๋‚ฎ์€ ์ˆซ์ž vs ๋†’์€ ์ˆซ์ž ๋น„์œจ, ์›”๋ณ„ ๋‹น์ฒจ ๋ฒˆํ˜ธ ๋ถ„์„ ๋“ฑ์„ ์ง„ํ–‰ํ•˜๊ณ , ๋จธ์‹ ๋Ÿฌ๋‹ ๋ชจ๋ธ์„ ํ™œ์šฉํ•ด ๋ฒˆํ˜ธ๋ฅผ ์˜ˆ์ธกํ•ด๋ณผ ๊ฒƒ์ด๋‹ค.1. ํ™€์ˆ˜ vs ์ง์ˆ˜ and ๋‚ฎ์€ ์ˆซ์ž(1~22) vs ๋†’์€ ์ˆซ์ž(23~45) ๋น„์œจ ๋ถ„์„๋กœ๋˜ ๋‹น์ฒจ ๋ฒˆํ˜ธ์—์„œ ํ™€์ˆ˜์™€ ์ง์ˆ˜์˜ ์ถœํ˜„ ๋น„์œจ์„ ๋ถ„์„ํ•œ๋‹ค.def odd_eve..

[Python][AI] RAG (Retrieval-Augmented Generation)๋ž€ ๋ฌด์—‡์ธ๊ฐ€?

์ตœ๊ทผ ์ธ๊ณต์ง€๋Šฅ(AI) ๋ชจ๋ธ์ด ๋ฐœ์ „ํ•˜๋ฉด์„œ ๋Œ€ํ™”ํ˜• AI, ์ฑ—๋ด‡, ๋ฌธ์„œ ์š”์•ฝ, ์ฝ”๋“œ ์ƒ์„ฑ ๋“ฑ ๋‹ค์–‘ํ•œ ๋ถ„์•ผ์—์„œ ์ž์—ฐ์–ด ์ฒ˜๋ฆฌ(NLP) ๊ธฐ์ˆ ์ด ํ™œ์šฉ๋˜๊ณ  ์žˆ๋‹ค. ํŠนํžˆ, ๋Œ€ํ˜• ์–ธ์–ด ๋ชจ๋ธ(LLM, Large Language Model)์˜ ์„ฑ๋Šฅ์ด ๊ธ‰๊ฒฉํžˆ ํ–ฅ์ƒ๋˜๋ฉด์„œ ์ƒ์„ฑํ˜• AI(Generative AI)์˜ ๊ฐ€๋Šฅ์„ฑ์ด ๋”์šฑ ํ™•๋Œ€๋˜๊ณ  ์žˆ๋‹ค. ํ•˜์ง€๋งŒ ์ด๋Ÿฌํ•œ ๋ชจ๋ธ์—๋„ ๋ช‡ ๊ฐ€์ง€ ํ•œ๊ณ„๊ฐ€ ์žˆ๋‹ค.์‚ฌ์ „ ํ•™์Šต ๋ฐ์ดํ„ฐ์— ์˜์กด์ ์ž„ – LLM์€ ํ•™์Šต๋œ ๋ฐ์ดํ„ฐ ๋‚ด์—์„œ๋งŒ ์ •๋ณด๋ฅผ ์ƒ์„ฑํ•  ์ˆ˜ ์žˆ์Œ. ์ตœ์‹  ์ •๋ณด๋‚˜ ์™ธ๋ถ€ ๋ฌธ์„œ๋ฅผ ์ฆ‰๊ฐ ๋ฐ˜์˜ํ•˜๋Š” ๊ฒƒ์ด ์–ด๋ ค์›€ํ™˜๊ฐ(Hallucination) ๋ฌธ์ œ – ์กด์žฌํ•˜์ง€ ์•Š๋Š” ์ •๋ณด๋ฅผ ์ƒ์„ฑํ•  ๊ฐ€๋Šฅ์„ฑ์ด ์žˆ์Œ์ •ํ™•์„ฑ ๋ฐ ์‹ ๋ขฐ์„ฑ ๋ถ€์กฑ – ๋ชจ๋ธ์ด ์‚ฌ์‹ค๊ณผ ๋‹ค๋ฅธ ๋‚ด์šฉ์„ ํฌํ•จํ•  ์ˆ˜ ์žˆ์Œ์ด๋Ÿฌํ•œ ๋ฌธ์ œ๋ฅผ ํ•ด๊ฒฐํ•˜๊ธฐ ์œ„ํ•ด ๋“ฑ์žฅํ•œ ๊ธฐ์ˆ ์ด ๋ฐ”๋กœ RAG(Re..

[Python][AI] ํ•œ๊ตญ ๋กœ๋˜ ๋ถ„์„: ๋‹น์ฒจ ๋ฒˆํ˜ธ ๋ถ„์„๊ณผ ํŒจํ„ด ์ฐพ๊ธฐ(EDA)

2025.02.18 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] ํ•œ๊ตญ ๋กœ๋˜ ๋ถ„์„: ๋‹น์ฒจ ํ™•๋ฅ ๊ณผ ์˜ˆ์ธก์˜ ๋ถˆ๊ฐ€๋Šฅ์„ฑ [Python][AI] ํ•œ๊ตญ ๋กœ๋˜ ๋ถ„์„: ๋‹น์ฒจ ํ™•๋ฅ ๊ณผ ์˜ˆ์ธก์˜ ๋ถˆ๊ฐ€๋Šฅ์„ฑ1. ํ•œ๊ตญ ๋กœ๋˜ ๊ฐœ๋…๋กœ๋˜๋Š” 1๋ถ€ํ„ฐ 45๊นŒ์ง€์˜ ์ˆซ์ž ์ค‘ 6๊ฐœ๋ฅผ ์„ ํƒํ•˜๋Š” ๋ฐฉ์‹์˜ ๋ณต๊ถŒ์ด๋‹ค. ๊ตฌ๋งค์ž๋Š” ๋‹ค์Œ๊ณผ ๊ฐ™์€ ๋ฐฉ๋ฒ•์œผ๋กœ ๋ฒˆํ˜ธ๋ฅผ ์„ ํƒํ•  ์ˆ˜ ์žˆ๋‹ค.์ž๋™ ์„ ํƒ: 45๊ฐœ์˜ ๋ฒˆํ˜ธ ์ค‘ 6๊ฐœ ๋ฒˆํ˜ธ๋ฅผ ๋ฌด์ž‘์œ„๋กœ ๋ถ€์—ฌ๋ฐ›5hr1rnp.tistory.com ๋ณธ ๊ธ€์—์„œ๋Š” ๋กœ๋˜ ๋‹น์ฒจ ๋ฐ์ดํ„ฐ์— ๋Œ€ํ•œ ํƒ์ƒ‰์  ๋ฐ์ดํ„ฐ ๋ถ„์„(EDA)์„ ์ง„ํ–‰ํ•˜๋ฉฐ, ๋‹ค์–‘ํ•œ ํ†ต๊ณ„์  ํŠน์ง•์„ ์‚ดํŽด๋ณด๊ฒ ๋‹ค.1. ๋ฐ์ดํ„ฐ ์ „์ฒ˜๋ฆฌ๋กœ๋˜ ๋ฐ์ดํ„ฐ๋Š” ๋™ํ–‰๋ณต๊ถŒ ์‚ฌ์ดํŠธ์—์„œ ์ œ๊ณตํ•˜๋Š” ์—‘์…€ ํŒŒ์ผ์„ ๊ฐ๊ฐ lotto-1.xls(1~600ํšŒ), lotto-2.xls(601~1,159ํšŒ)๋กœ ์ €์žฅํ•จ..

[Python][AI] ํ•œ๊ตญ ๋กœ๋˜ ๋ถ„์„: ๋‹น์ฒจ ํ™•๋ฅ ๊ณผ ์˜ˆ์ธก์˜ ๋ถˆ๊ฐ€๋Šฅ์„ฑ

1. ํ•œ๊ตญ ๋กœ๋˜ ๊ฐœ๋…๋กœ๋˜๋Š” 1๋ถ€ํ„ฐ 45๊นŒ์ง€์˜ ์ˆซ์ž ์ค‘ 6๊ฐœ๋ฅผ ์„ ํƒํ•˜๋Š” ๋ฐฉ์‹์˜ ๋ณต๊ถŒ์ด๋‹ค. ๊ตฌ๋งค์ž๋Š” ๋‹ค์Œ๊ณผ ๊ฐ™์€ ๋ฐฉ๋ฒ•์œผ๋กœ ๋ฒˆํ˜ธ๋ฅผ ์„ ํƒํ•  ์ˆ˜ ์žˆ๋‹ค.์ž๋™ ์„ ํƒ: 45๊ฐœ์˜ ๋ฒˆํ˜ธ ์ค‘ 6๊ฐœ ๋ฒˆํ˜ธ๋ฅผ ๋ฌด์ž‘์œ„๋กœ ๋ถ€์—ฌ๋ฐ›์Œ๋ฐ˜์ž๋™ ์„ ํƒ: ์›ํ•˜๋Š” ๋ฒˆํ˜ธ ์ผ๋ถ€(1~5๊ฐœ)๋ฅผ ์ง์ ‘ ์„ ํƒํ•˜๊ณ , ๋‚˜๋จธ์ง€๋Š” ๋ฌด์ž‘์œ„๋กœ ์ฑ„์›€์ˆ˜๋™ ์„ ํƒ: 6๊ฐœ ๋ฒˆํ˜ธ๋ฅผ ์ „๋ถ€ ์ง์ ‘ ์„ ํƒ์ถ”์ฒจ์€ ๋งค์ฃผ ํ† ์š”์ผ ์ €๋…์— ์ง„ํ–‰๋˜๋ฉฐ, 6๊ฐœ์˜ ๋‹น์ฒจ๋ฒˆํ˜ธ์™€ 1๊ฐœ์˜ ๋ณด๋„ˆ์Šค๋ฒˆํ˜ธ๊ฐ€ ๋ฌด์ž‘์œ„๋กœ ๊ฒฐ์ •๋œ๋‹ค. ๋ณด๋„ˆ์Šค๋ฒˆํ˜ธ๋Š” 2๋“ฑ ๋‹น์ฒจ์ž๋ฅผ ํŒ๋ณ„ํ•˜๋Š” ๋ฐ ์‚ฌ์šฉ๋œ๋‹ค.2. ๋‹น์ฒจ ๋“ฑ๊ธ‰๊ณผ ํ™•๋ฅ ๋กœ๋˜์˜ ๋‹น์ฒจ์€ ๋ฒˆํ˜ธ ์ผ์น˜ ๊ฐœ์ˆ˜์— ๋”ฐ๋ผ 5๊ฐœ์˜ ๋“ฑ๊ธ‰์œผ๋กœ ๋‚˜๋‰จ. ๊ฐ ๋“ฑ๊ธ‰๋ณ„ ๋‹น์ฒจ ํ™•๋ฅ ๊ณผ ๋‹น์ฒจ๊ธˆ ๋ฐฐ๋ถ„ ๊ตฌ์กฐ๋Š” ๋‹ค์Œ๊ณผ ๊ฐ™์Œ. ๋“ฑ์œ„๋‹น์ฒจ ์กฐ๊ฑด๋‹น์ฒจ ํ™•๋ฅ ๋‹น์ฒจ๊ธˆ ๋ฐฐ๋ถ„1๋“ฑ6๊ฐœ ๋ฒˆํ˜ธ ์ผ์น˜1 / 8,145,060์ด ๋‹น์ฒจ๊ธˆ ์ค‘ 4๋“ฑ, 5๋“ฑ ์ œ์™ธ ๊ธˆ..

[Python][AI] OpenCV YuNet์„ ํ™œ์šฉํ•œ ๋™์˜์ƒ ์–ผ๊ตด ๋ชจ์ž์ดํฌ ์ฒ˜๋ฆฌ

2025.02.17 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] OpenCV: ์ปดํ“จํ„ฐ ๋น„์ „ ๋ผ์ด๋ธŒ๋Ÿฌ๋ฆฌ ์†Œ๊ฐœ2025.02.17 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] OpenCV YuNet์„ ํ™œ์šฉํ•œ ์–ผ๊ตด ํƒ์ง€2025.02.17 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] OpenCV YuNet์„ ํ™œ์šฉํ•œ ์–ผ๊ตด ๋ชจ์ž์ดํฌ ์ฒ˜๋ฆฌ 1. ๋™์˜์ƒ ์–ผ๊ตด ๋ชจ์ž์ดํฌ ๊ฐœ์š”์ด์ „ ๊ธ€์—์„œ๋Š” ์ •์ ์ธ ์ด๋ฏธ์ง€ ๋˜๋Š” ๋””๋ ‰ํ† ๋ฆฌ์—์„œ ์–ผ๊ตด์„ ํƒ์ง€ํ•˜๊ณ  ๋ชจ์ž์ดํฌ๋ฅผ ์ ์šฉํ•˜๋Š” ๋ฐฉ๋ฒ•์„ ๋‹ค๋ฃจ์—ˆ๋‹ค. ์ด๋ฒˆ ๊ธ€์—์„œ๋Š” ๋™์˜์ƒ์—์„œ ์–ผ๊ตด์„ ์ž๋™์œผ๋กœ ํƒ์ง€ํ•˜๊ณ  ๋ชจ์ž์ดํฌ๋ฅผ ์ ์šฉํ•˜๋Š” ๋ฐฉ๋ฒ•์„ ๋‹ค๋ฃฌ๋‹ค.OpenCV์˜ YuNet ์–ผ๊ตด ํƒ์ง€ ๋ชจ๋ธ์„ ํ™œ์šฉํ•˜์—ฌ ์‹ค์‹œ๊ฐ„์œผ๋กœ ์–ผ๊ตด์„ ๊ฐ์ง€ํ•˜๊ณ , ํ•ด๋‹น ์˜์—ญ์„ ๋ชจ์ž์ดํฌ ์ฒ˜๋ฆฌ..

[Python][AI] OpenCV YuNet์„ ํ™œ์šฉํ•œ ์–ผ๊ตด ๋ชจ์ž์ดํฌ ์ฒ˜๋ฆฌ

2025.02.17 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] OpenCV: ์ปดํ“จํ„ฐ ๋น„์ „ ๋ผ์ด๋ธŒ๋Ÿฌ๋ฆฌ ์†Œ๊ฐœ2025.02.17 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] OpenCV YuNet์„ ํ™œ์šฉํ•œ ์–ผ๊ตด ํƒ์ง€ 1. ์–ผ๊ตด ๋ชจ์ž์ดํฌ ์ฒ˜๋ฆฌ ๊ฐœ์š”์ด๋ฒˆ ๊ธ€์—์„œ๋Š” OpenCV์˜ YuNet ๋ชจ๋ธ์„ ์ด์šฉํ•˜์—ฌ ์ด๋ฏธ์ง€์—์„œ ์–ผ๊ตด์„ ์ž๋™์œผ๋กœ ํƒ์ง€ํ•˜๊ณ , ํ•ด๋‹น ์˜์—ญ์— ๋ชจ์ž์ดํฌ๋ฅผ ์ ์šฉํ•˜๋Š” ๋ฐฉ๋ฒ•์„ ๋‹ค๋ฃฐ ๊ฒƒ์ด๋‹ค. 2. ์‹คํ–‰ ํ™˜๊ฒฝ ์ค€๋น„OpenCV ์„ค์น˜pip install opencv-python 3. YuNet ์–ผ๊ตด ํƒ์ง€ ๋ชจ๋ธ ๋‹ค์šด๋กœ๋“œ import osimport requests# YuNet ๋ชจ๋ธ ํŒŒ์ผ ์„ค์ •model_filename = "face_detection_yunet_2023mar.o..

[Python][AI] OpenCV YuNet์„ ํ™œ์šฉํ•œ ์–ผ๊ตด ํƒ์ง€

2025.02.17 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] OpenCV: ์ปดํ“จํ„ฐ ๋น„์ „ ๋ผ์ด๋ธŒ๋Ÿฌ๋ฆฌ ์†Œ๊ฐœ1. YuNet์„ ์ด์šฉํ•œ ์–ผ๊ตด ํƒ์ง€ ๊ฐœ์š”OpenCV๋Š” ๊ฐ•๋ ฅํ•œ ์ด๋ฏธ์ง€ ์ฒ˜๋ฆฌ ๊ธฐ๋Šฅ์„ ์ œ๊ณตํ•˜๋ฉฐ, cv2.FaceDetectorYN์„ ์ด์šฉํ•˜๋ฉด ๋”ฅ๋Ÿฌ๋‹ ๊ธฐ๋ฐ˜ ์–ผ๊ตด ํƒ์ง€๋ฅผ ๊ฐ„๋‹จํ•˜๊ฒŒ ์ˆ˜ํ–‰ํ•  ์ˆ˜ ์žˆ๋‹ค. ์ด๋ฒˆ ๊ธ€์—์„œ๋Š” OpenCV์˜ YuNet ๋ชจ๋ธ์„ ์‚ฌ์šฉํ•˜์—ฌ ์ด๋ฏธ์ง€์—์„œ ์–ผ๊ตด์„ ํƒ์ง€ํ•˜๋Š” ๋ฐฉ๋ฒ•์„ ๋‹ค๋ฃฌ๋‹ค.2. ์‹คํ–‰ ํ™˜๊ฒฝ ์ค€๋น„(1) OpenCV ์„ค์น˜YuNet์„ ์‚ฌ์šฉํ•˜๋ ค๋ฉด opencv-python ํŒจํ‚ค์ง€๋งŒ ์„ค์น˜ํ•˜๋ฉด ๋จ. Jupyter Notebook์—์„œ ๋‹ค์Œ ๋ช…๋ น์–ด๋ฅผ ์‹คํ–‰ํ•˜์—ฌ ์„ค์น˜ํ•  ์ˆ˜ ์žˆ์Œ.pip install opencv-python ์„ค์น˜๊ฐ€ ์™„๋ฃŒ๋˜๋ฉด, OpenCV ๋ฒ„์ „์„ ํ™•์ธํ•˜์—ฌ ์ •์ƒ์ ์œผ๋กœ ์„ค์น˜๋˜์—ˆ๋Š”..

[Python][AI] OpenCV: ์ปดํ“จํ„ฐ ๋น„์ „ ๋ผ์ด๋ธŒ๋Ÿฌ๋ฆฌ ์†Œ๊ฐœ

1. OpenCV๋ž€?OpenCV(Open Source Computer Vision)๋Š” ์‹ค์‹œ๊ฐ„ ์ด๋ฏธ์ง€ ์ฒ˜๋ฆฌ ๋ฐ ์ปดํ“จํ„ฐ ๋น„์ „ ์• ํ”Œ๋ฆฌ์ผ€์ด์…˜์„ ์œ„ํ•œ ์˜คํ”ˆ์†Œ์Šค ๋ผ์ด๋ธŒ๋Ÿฌ๋ฆฌ. ์ธํ…”์—์„œ ๊ฐœ๋ฐœํ•˜๊ณ  2000๋…„ 6์›”์— ๋ฐœํ‘œ๋˜์—ˆ์œผ๋ฉฐ, ํ˜„์žฌ๋Š” ์ธํ…”์ด ์†์„ ๋—€ ์ƒํƒœ๋‹ค. Python, C++, Java ๋“ฑ์„ ์ง€์›ํ•˜๋ฉฐ, ๋‹ค์–‘ํ•œ ์šด์˜์ฒด์ œ(Windows, Linux, macOS)์—์„œ ์‹คํ–‰ ๊ฐ€๋Šฅํ•˜๋‹ค.2. OpenCV์˜ ์ฃผ์š” ๊ธฐ๋ŠฅOpenCV๋Š” ์ปดํ“จํ„ฐ ๋น„์ „ ๋ฐ ์ด๋ฏธ์ง€ ์ฒ˜๋ฆฌ์™€ ๊ด€๋ จ๋œ ๋‹ค์–‘ํ•œ ๊ธฐ๋Šฅ์„ ์ œ๊ณตํ•จ.(1) ์ด๋ฏธ์ง€ ๋ฐ ๋™์˜์ƒ ์ฒ˜๋ฆฌ์ด๋ฏธ์ง€ ์ฝ๊ธฐ, ์ €์žฅ, ๋ณ€ํ™˜๋™์˜์ƒ ์žฌ์ƒ ๋ฐ ํ”„๋ ˆ์ž„ ์ฒ˜๋ฆฌ์ƒ‰๊ณต๊ฐ„ ๋ณ€ํ™˜ (RGB ↔ Grayscale, HSV ๋“ฑ)(2) ํ•„ํ„ฐ ๋ฐ ๋ณ€ํ™˜๋ธ”๋Ÿฌ๋ง ๋ฐ ์ƒคํ”„๋‹์—ฃ์ง€ ๊ฒ€์ถœ (Canny Edge Detection)๊ธฐํ•˜ํ•™์  ๋ณ€ํ™˜ (ํšŒ์ „..

[Python][AI] ์„ ํ˜• vs ๋น„์„ ํ˜•, ๋‹จ์กฐ vs ๋น„๋‹จ์กฐ ๋ฐ์ดํ„ฐ์…‹ ์ •๋ฆฌ

๋ฐ์ดํ„ฐ ๋ถ„์„๊ณผ ๋จธ์‹ ๋Ÿฌ๋‹์—์„œ ๋ฐ์ดํ„ฐ๋ฅผ ์ดํ•ดํ•˜๋Š” ๊ฒƒ์€ ๋งค์šฐ ์ค‘์š”ํ•˜๋‹ค. ํŠนํžˆ ๋ฐ์ดํ„ฐ์˜ ๊ตฌ์กฐ๋ฅผ ํŒŒ์•…ํ•˜๋Š” ๊ฒƒ์€ ๋ชจ๋ธ ์„ ํƒ๊ณผ ํ•ด์„์— ํฐ ์˜ํ–ฅ์„ ๋ฏธ์นœ๋‹ค. ๋ฐ์ดํ„ฐ๋Š” ํฌ๊ฒŒ ์„ ํ˜•(linear)๊ณผ ๋น„์„ ํ˜•(non-linear), ๋‹จ์กฐ(monotonic)์™€ ๋น„๋‹จ์กฐ(non-monotonic) ํ˜•ํƒœ๋กœ ๊ตฌ๋ถ„ํ•  ์ˆ˜ ์žˆ๋‹ค.1. ์„ ํ˜•(Linear) ๋ฐ์ดํ„ฐ vs ๋น„์„ ํ˜•(Non-Linear) ๋ฐ์ดํ„ฐ(1) ์„ ํ˜• ๋ฐ์ดํ„ฐ (Linear Data)์„ ํ˜• ๋ฐ์ดํ„ฐ๋Š” ๋ณ€์ˆ˜ ๊ฐ„์˜ ๊ด€๊ณ„๊ฐ€ ์ง์„ ์œผ๋กœ ํ‘œํ˜„๋  ์ˆ˜ ์žˆ๋Š” ๋ฐ์ดํ„ฐ๋ฅผ ์˜๋ฏธํ•จ. ์ฆ‰, ํ•œ ๋ณ€์ˆ˜๊ฐ€ ์ฆ๊ฐ€ํ•  ๋•Œ ๋‹ค๋ฅธ ๋ณ€์ˆ˜๋„ ์ผ์ •ํ•œ ๋น„์œจ๋กœ ์ฆ๊ฐ€ํ•˜๊ฑฐ๋‚˜ ๊ฐ์†Œํ•˜๋Š” ํŒจํ„ด์„ ๋ณด์ž„.ํŠน์ง•๋ฐ์ดํ„ฐ๊ฐ€ ์ง์„  ๊ด€๊ณ„๋ฅผ ๋”ฐ๋ฆ„์„ ํ˜• ํšŒ๊ท€(Linear Regression) ๋ชจ๋ธ๋กœ ์‰ฝ๊ฒŒ ์„ค๋ช… ๊ฐ€๋Šฅ์˜ˆ์ธก์ด ์ง๊ด€์ ์ด๋ฉฐ ๊ณ„์‚ฐ ๋น„์šฉ์ด ๋‚ฎ์Œ์˜ˆ์ œํ‚ค์™€ ๋ชธ๋ฌด๊ฒŒ๊ฐ€..

[Python][AI] AI ๋ชจ๋ธ ์„ฑ๋Šฅ ์ง€ํ‘œ(Metrics) ์ •๋ฆฌ

AI ๋ชจ๋ธ์˜ ์„ฑ๋Šฅ์„ ํ‰๊ฐ€ํ•˜๊ธฐ ์œ„ํ•ด ๋‹ค์–‘ํ•œ ์„ฑ๋Šฅ ์ง€ํ‘œ(metrics)๊ฐ€ ์‚ฌ์šฉ๋œ๋‹ค. ๋ชจ๋ธ์ด ์–ผ๋งˆ๋‚˜ ์ •ํ™•ํ•˜๊ฒŒ ์ž‘๋™ํ•˜๋Š”์ง€, ์˜ˆ์ธก์ด ์–ผ๋งˆ๋‚˜ ์‹ ๋ขฐํ•  ์ˆ˜ ์žˆ๋Š”์ง€๋ฅผ ํ‰๊ฐ€ํ•˜๋Š” ๋ฐ ์ค‘์š”ํ•œ ์—ญํ• ์„ ํ•˜๋‹ค. ์ด๋ฒˆ ๊ธ€์—์„œ ๋Œ€ํ‘œ์ ์ธ ์„ฑ๋Šฅ ์ง€ํ‘œ๋“ค์„ ์†Œ๊ฐœํ•˜๊ฒ ๋‹ค.1. ํšŒ๊ท€(Regression) ๋ชจ๋ธ ์„ฑ๋Šฅ ์ง€ํ‘œํšŒ๊ท€ ๋ชจ๋ธ์€ ์—ฐ์†์ ์ธ ์ˆซ์ž ๊ฐ’์„ ์˜ˆ์ธกํ•˜๋Š” ๋ชจ๋ธ์ด๋ฉฐ, ์ฃผ๋กœ ํ‰๊ท  ์ œ๊ณฑ ์˜ค์ฐจ(Mean Squared Error)๋‚˜ ํ‰๊ท  ์ ˆ๋Œ€ ์˜ค์ฐจ(Mean Absolute Error) ๋“ฑ์˜ ์ง€ํ‘œ๋ฅผ ์‚ฌ์šฉํ•จ.(1) MSE (Mean Squared Error, ํ‰๊ท  ์ œ๊ณฑ ์˜ค์ฐจ)MSE๋Š” ์‹ค์ œ ๊ฐ’๊ณผ ์˜ˆ์ธก ๊ฐ’์˜ ์ฐจ์ด๋ฅผ ์ œ๊ณฑํ•œ ํ›„ ํ‰๊ท ์„ ๊ตฌํ•œ ๊ฐ’์ž„. ๊ฐ’์ด ์ž‘์„์ˆ˜๋ก ๋ชจ๋ธ์˜ ์˜ˆ์ธก์ด ์‹ค์ œ ๊ฐ’๊ณผ ๊ฐ€๊นŒ์›€์„ ์˜๋ฏธํ•จ.์žฅ์ : ์˜ค์ฐจ๋ฅผ ์ œ๊ณฑํ•˜์—ฌ ํฐ ์˜ค์ฐจ๋ฅผ ๊ฐ•์กฐํ•จ๋‹จ์ : ์ œ๊ณฑ์ด ์ ์šฉ๋˜..

[AI][CatBoost] CatBoost๋กœ Wine Quality ์˜ˆ์ธกํ•˜๊ธฐ

2025.01.23 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] ํƒ์ƒ‰์  ๋ฐ์ดํ„ฐ ๋ถ„์„(EDA) - ์™€์ธ ํ’ˆ์งˆ ๋ฐ์ดํ„ฐ์…‹ (Wine Quality Dataset) - 12025.01.24 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] ํƒ์ƒ‰์  ๋ฐ์ดํ„ฐ ๋ถ„์„(EDA) - ์™€์ธ ํ’ˆ์งˆ ๋ฐ์ดํ„ฐ์…‹ (Wine Quality Dataset) - 22025.02.04 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] ํƒ์ƒ‰์  ๋ฐ์ดํ„ฐ ๋ถ„์„(EDA) - ์™€์ธ ํ’ˆ์งˆ ๋ฐ์ดํ„ฐ์…‹ (Wine Quality Dataset) - 32025.02.04 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] ํƒ์ƒ‰์  ๋ฐ์ดํ„ฐ ๋ถ„์„(EDA) - ์™€์ธ ํ’ˆ์งˆ ๋ฐ์ดํ„ฐ์…‹ (Wine Qualit..

[Python][AI] CatBoost: ๊ฐ•๋ ฅํ•œ ๊ทธ๋ž˜๋””์–ธํŠธ ๋ถ€์ŠคํŒ… ๋ผ์ด๋ธŒ๋Ÿฌ๋ฆฌ

1. CatBoost๋ž€?CatBoost๋Š” Yandex์—์„œ ๊ฐœ๋ฐœํ•œ ๊ทธ๋ž˜๋””์–ธํŠธ ๋ถ€์ŠคํŒ…(Gradient Boosting) ๊ธฐ๋ฐ˜ ๋จธ์‹ ๋Ÿฌ๋‹ ๋ผ์ด๋ธŒ๋Ÿฌ๋ฆฌ๋กœ, ์˜์‚ฌ๊ฒฐ์ • ํŠธ๋ฆฌ๋ฅผ ๊ธฐ๋ฐ˜์œผ๋กœ ํ•™์Šตํ•˜๋Š” ๋ฐฉ์‹์ด๋‹ค. ๊ณ ์„ฑ๋Šฅ ๋ฐ ๋†’์€ ์ •ํ™•๋„๋ฅผ ์ž๋ž‘ํ•˜๋ฉฐ, ์ถ”์ฒœ ์‹œ์Šคํ…œ, ๊ฒ€์ƒ‰ ์—”์ง„, ์ž์œจ ์ฃผํ–‰, ๋‚ ์”จ ์˜ˆ์ธก ๋“ฑ ๋‹ค์–‘ํ•œ ๋ถ„์•ผ์—์„œ ์‚ฌ์šฉ๋œ๋‹ค.2. CatBoost์˜ ์ฃผ์š” ํŠน์ง•1) ๋ณ„๋„ ํŒŒ๋ผ๋ฏธํ„ฐ ํŠœ๋‹ ์—†์ด๋„ ์šฐ์ˆ˜ํ•œ ์„ฑ๋Šฅ๊ธฐ๋ณธ ์„ค์ •๊ฐ’์œผ๋กœ๋„ ์ข‹์€ ์„ฑ๋Šฅ์„ ๋ฐœํœ˜ํ•จ์‚ฌ์šฉ์ž๊ฐ€ ๋ณต์žกํ•œ ํŒŒ๋ผ๋ฏธํ„ฐ ํŠœ๋‹์— ๋งŽ์€ ์‹œ๊ฐ„์„ ์†Œ๋น„ํ•˜์ง€ ์•Š์•„๋„ ๋จ2) ๋ฒ”์ฃผํ˜• ๋ฐ์ดํ„ฐ(Categorical Features) ์ง€์›์ผ๋ฐ˜์ ์ธ ๋จธ์‹ ๋Ÿฌ๋‹ ๋ชจ๋ธ์—์„œ๋Š” ๋ฒ”์ฃผํ˜• ๋ฐ์ดํ„ฐ๋ฅผ ์ˆซ์ž๋กœ ๋ณ€ํ™˜ํ•ด์•ผ ํ•˜์ง€๋งŒ, CatBoost๋Š” ์ž๋™์œผ๋กœ ๋ฒ”์ฃผํ˜• ๋ฐ์ดํ„ฐ๋ฅผ ์ฒ˜๋ฆฌํ•˜์—ฌ ๋ชจ๋ธ ์„ฑ๋Šฅ์„ ํ–ฅ์ƒ์‹œํ‚ด3) ๋น ๋ฅด๊ณ  ํ™•์žฅ ๊ฐ€๋Šฅํ•œ GPU..

[Python][AI] AutoGluon: ์‰ฝ๊ณ  ๋น ๋ฅธ ๋จธ์‹ ๋Ÿฌ๋‹ ์ž๋™ํ™” ๋ผ์ด๋ธŒ๋Ÿฌ๋ฆฌ

1. AutoGluon ๊ฐœ์š” ๋ฐ ์—ฐํ˜AutoGluon์€ AWS(Amazon Web Services)์—์„œ ๊ฐœ๋ฐœํ•œ AutoML(Automated Machine Learning) ๋ผ์ด๋ธŒ๋Ÿฌ๋ฆฌ๋กœ, ๋จธ์‹ ๋Ÿฌ๋‹ ๋ชจ๋ธ์„ ์‰ฝ๊ณ  ๋น ๋ฅด๊ฒŒ ํ•™์Šตํ•  ์ˆ˜ ์žˆ๋„๋ก ์ง€์›ํ•œ๋‹ค. 2019๋…„ ์ฒ˜์Œ ๊ณต๊ฐœ๋˜์—ˆ์œผ๋ฉฐ, ์ตœ์‹  SOTA(State-of-the-Art) ๋ชจ๋ธ์„ ์ž๋™์œผ๋กœ ํ™œ์šฉํ•˜์—ฌ ๋†’์€ ์„ฑ๋Šฅ์„ ์ œ๊ณตํ•˜๋Š” ๊ฒƒ์ด ํŠน์ง•์ด๋‹ค. ์ฃผ์š” ํŠน์ง•:3์ค„์˜ ์ฝ”๋“œ๋กœ ๋จธ์‹ ๋Ÿฌ๋‹ ๋ชจ๋ธ ๊ตฌ์ถ• ๊ฐ€๋Šฅ์ตœ์‹  ๋”ฅ๋Ÿฌ๋‹ ๋ฐ ๋จธ์‹ ๋Ÿฌ๋‹ ๊ธฐ๋ฒ• ์ž๋™ ์ ์šฉ๊ฐ„ํŽธํ•œ ๋ฐฐํฌ ๋ฐ ํ™•์žฅ์„ฑ ์ œ๊ณต๋‹ค์–‘ํ•œ ๋ฐ์ดํ„ฐ ์œ ํ˜• ์ง€์›(ํ‘œํ˜• ๋ฐ์ดํ„ฐ, ์ด๋ฏธ์ง€, ํ…์ŠคํŠธ, ์‹œ๊ณ„์—ด ๋“ฑ)2. AutoGluon์˜ ์ฃผ์š” ๊ธฐ๋ŠฅAutoGluon์€ ์—ฌ๋Ÿฌ ๊ฐ€์ง€ ํ•™์Šต ์œ ํ˜•์„ ์ง€์›ํ•˜๋ฉฐ, ๋ฐ์ดํ„ฐ ์œ ํ˜•์— ๋”ฐ๋ผ ์ตœ์ ์˜ ๋ชจ๋ธ์„ ์ž๋™์œผ๋กœ ์„ ํƒํ•˜์—ฌ ํ•™์Šตํ• ..

[Python][AI] Exploratory Data Analysis (EDA) - Wine Quality Dataset - 4

2025.02.10 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] Exploratory Data Analysis (EDA) - Wine Quality Dataset - 12025.02.10 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] Exploratory Data Analysis (EDA) - Wine Quality Dataset - 22025.02.10 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] Exploratory Data Analysis (EDA) - Wine Quality Dataset - 3 1. IntroductionIn this section, we will use the XGBoost regression model to pre..

[Python][AI] Exploratory Data Analysis (EDA) - Wine Quality Dataset - 3

2025.02.10 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] Exploratory Data Analysis (EDA) - Wine Quality Dataset - 12025.02.10 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] Exploratory Data Analysis (EDA) - Wine Quality Dataset - 2 In this section, we will visualize the relationships between variables and identify key patterns in the dataset.Wine Quality Distribution & Correlation Analysis# Library Version# pandas..

[Python][AI] Exploratory Data Analysis (EDA) - Wine Quality Dataset - 2

2025.02.10 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] Exploratory Data Analysis (EDA) - Wine Quality Dataset - 1 [Python][AI] Exploratory Data Analysis (EDA) - Wine Quality Dataset - 1Exploratory Data Analysis (EDA) is the first step in data analysis, where data is visually explored, summary statistics are examined, and patterns and characteristics of the dataset are identified. In this post, we will ..

[Python][AI] Exploratory Data Analysis (EDA) - Wine Quality Dataset - 1

Exploratory Data Analysis (EDA) is the first step in data analysis, where data is visually explored, summary statistics are examined, and patterns and characteristics of the dataset are identified. In this post, we will walk through the step-by-step process of exploring data using the Wine Quality Dataset.What is EDA?EDA (Exploratory Data Analysis) is a crucial process for gaining a deeper under..

[Python][AI] ํƒ์ƒ‰์  ๋ฐ์ดํ„ฐ ๋ถ„์„(EDA) - ์™€์ธ ํ’ˆ์งˆ ๋ฐ์ดํ„ฐ์…‹ (Wine Quality Dataset) - 4

2025.01.23 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] ํƒ์ƒ‰์  ๋ฐ์ดํ„ฐ ๋ถ„์„(EDA) - ์™€์ธ ํ’ˆ์งˆ ๋ฐ์ดํ„ฐ์…‹ (Wine Quality Dataset) - 12025.01.24 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] ํƒ์ƒ‰์  ๋ฐ์ดํ„ฐ ๋ถ„์„(EDA) - ์™€์ธ ํ’ˆ์งˆ ๋ฐ์ดํ„ฐ์…‹ (Wine Quality Dataset) - 22025.02.04 - [๊ฐœ๋ฐœ Code/์ธ๊ณต์ง€๋Šฅ A.I.] - [Python][AI] ํƒ์ƒ‰์  ๋ฐ์ดํ„ฐ ๋ถ„์„(EDA) - ์™€์ธ ํ’ˆ์งˆ ๋ฐ์ดํ„ฐ์…‹ (Wine Quality Dataset) - 31. ๋“ค์–ด๊ฐ€๋ฉฐ์ด๋ฒˆ ๊ธ€์—์„œ๋Š” XGBoost ํšŒ๊ท€ ๋ชจ๋ธ์„ ํ™œ์šฉํ•˜์—ฌ ์™€์ธ์˜ ํ’ˆ์งˆ์„ ์˜ˆ์ธกํ•˜๊ณ , ๋‹ค์–‘ํ•œ ๊ธฐ๋ฒ•์„ ์ ์šฉํ•˜์—ฌ ์„ฑ๋Šฅ์„ ๋น„๊ตํ•ด๋ณด๊ฒ ๋‹ค.๋‹จ์ˆœํ•œ XGBoost ๋ชจ๋ธ์„ ..

728x90
๋ฐ˜์‘ํ˜•