๋ณธ๋ฌธ ๋ฐ”๋กœ๊ฐ€๊ธฐ

์ „์ฒด ๊ธ€

(154)
Asterisk @Asterisk(*) -๋‹จ์ˆœ ๊ณฑ์…ˆ, ์ œ๊ณฑ ์—ฐ์‚ฐ, ๊ฐ€๋ณ€ ์ธ์ž ํ™œ์šฉ ๋“ฑ ๋‹ค์–‘ํ•˜๊ฒŒ ์‚ฌ์šฉ๋จ **: ํ‚ค์›Œ๋“œ ์ธ์ž dict ํƒ€์ž…์œผ๋กœ ๋“ค์–ด๊ฐ ์‹ค์Šต) def asterisk_test(a, *args): print(a, args) # 1 (2, 3, 4, 5, 6) print(type(args)) # asterisk_test(1,2,3,4,5,6) def asterisk_test(a, **kargs): print(a, kargs) # 1 {'b': 2, 'c': 3, 'd': 4, 'e': 5, 'f': 6} print(type(kargs)) # asterisk_test(1, b=2, c=3, d=4, e=5, f=6) kargs: ํ‚ค์›Œ๋“œ ์ธ์ž ์˜ˆ์‹œ์—์„œ 1์€ a์— ๋“ค์–ด๊ฐ€๊ณ  ๋‚˜๋จธ์ง€๋Š” dict ํƒ€์ž…์œผ๋กœ kargs์— ๋“ค์–ด๊ฐ..
Lambda & MapReduce @Lambda -ํ•จ์ˆ˜ ์ด๋ฆ„์—†์ด ํ•จ์ˆ˜์ฒ˜๋Ÿผ ์“ธ ์ˆ˜ ์žˆ๋Š” ์ต๋ช… ํ•จ์ˆ˜ -์ˆ˜ํ•™์˜ ๋žŒ๋‹ค ๋Œ€์ˆ˜์—์„œ ์œ ๋ž˜ -Python3๋ถ€ํ„ฐ๋Š” ๊ถŒ์žฅํ•˜์ง€๋Š” ์•Š์ง€๋งŒ ์—ฌ์ „ํžˆ ๋งŽ์ด ์“ฐ์ž„ (List Comprehension์œผ๋กœ ๋Œ€์ฒด ๊ฐ€๋Šฅํ•˜์ง€๋งŒ ํŒ๋‹ค์Šค ๋“ฑ์—์„œ ํšจ์œจ์ ์œผ๋กœ ์‚ฌ์šฉ ๊ฐ€๋Šฅํ•˜๊ธฐ ๋•Œ๋ฌธ) ๋ณดํ†ต ํ•จ์ˆ˜์—์„œ ํ•จ์ˆ˜ ์„ ์–ธ ํ›„ ํ•จ์ˆ˜ ์ด๋ฆ„ ์ง€์ •, ์ธ์ž๊ฐ’ ์ง€์ •, ๋ฆฌํ„ด ๊ฐ’ ์„ ์–ธํ•ด์ฃผ๋Š” ํ˜•ํƒœ๋ผ๋ฉด ๋žŒ๋‹ค ํ•จ์ˆ˜๋Š” ๋žŒ๋‹ค ์„ ์–ธ ํ›„ ์ž…๋ ฅ๊ฐ’, ๋ฐ˜ํ™˜๊ฐ’ ์„ ์–ธํ•ด์ฃผ๋Š” ํ˜•ํƒœ (์ต๋ช… ํ•จ์ˆ˜๋ฅผ ๋ณ€์ˆ˜์— ๋„ฃ์–ด์ฃผ๋ฉด ๊ทธ ๋ณ€์ˆ˜ ์ž์ฒด๊ฐ€ ํ•˜๋‚˜์˜ ํ•จ์ˆ˜๊ฐ€ ๋˜๋Š” ํ˜•ํƒœ) @Map ํ•จ์ˆ˜ -Squence ์ž๋ฃŒํ˜•(List, Tuple) ๊ฐ element์— ๋™์ผํ•œ function ์ ์šฉ -์ฝ”๋“œ ์ง๊ด€์„ฑ์ด ๋–จ์–ด์ง€๊ธฐ ๋•Œ๋ฌธ์— Python3๋ถ€ํ„ฐ๋Š” ๊ถŒ์žฅํ•˜์ง€๋Š” ์•Š์ง€๋งŒ ์—ฌ์ „ํžˆ ๋งŽ์ด ์“ฐ์ž„ ์‹ค์Šต) # 1 def f(x, y): ret..
Enumerate & Zip @Enumerate -๋ฆฌ์ŠคํŠธ์—์„œ ๊ฐ’ ์ถ”์ถœํ•  ๋•Œ ์ธ๋ฑ์Šค ๋ฒˆํ˜ธ๋ฅผ ๊ฐ™์ด ์ถ”์ถœํ•˜๋Š” ํ•จ์ˆ˜ ์‹ค์Šต) # Unpacking index and value in list for i, v in enumerate(['tic', 'tac', 'tok']): print(i, v) # 0 tic # 1 tac # 2 tok # Unpacking the index and value in the list and saving it as a list mylist = ['a', 'b', 'c', 'd'] print(list(enumerate(mylist))) # [(0, 'a'), (1, 'b'), (2, 'c'), (3, 'd')] # Make a sentence into a list, unpack the list index and va..
List Comprehension @List comprehension -๊ธฐ์กด List ์‚ฌ์šฉํ•ด ๊ฐ„๋‹จํžˆ ๋‹ค๋ฅธ List ๋งŒ๋“œ๋Š” ๊ธฐ๋ฒ• -ํฌ๊ด„์ ์ธ List, ํฌํ•จ๋˜๋Š” ๋ฆฌ์ŠคํŠธ๋ผ๋Š” ์˜๋ฏธ๋กœ ์‚ฌ์šฉ -ํŒŒ์ด์ฌ์—์„œ ๊ฐ€์žฅ ๋งŽ์ด ์‚ฌ์šฉ๋˜๋Š” ๊ธฐ๋ฒ• ์ค‘ ํ•˜๋‚˜ -์ผ๋ฐ˜์ ์œผ๋กœ for + append ๋ณด๋‹ค ์†๋„ ๋น ๋ฆ„ ์‹ค์Šต) # for loop + append result = [] # create a list for i in range(10): result.append(i) print(result) # list Comprehension result = [i for i in range(10)] print(result) # print even numbers only result = [i for i in range(10) if i % 2 == 0] print(result) # Nes..
Split & Join @Split ํ•จ์ˆ˜ -String Type์˜ ๊ฐ’์„ ๋‚˜๋ˆ  List ํ˜•ํƒœ๋กœ ๋ฐ˜ํ™˜ -์ž๋ฅด๋Š” ๋ฐ์— ๋ชฉ์  ์žˆ์Œ -๋‰ด์Šค๋ฐ์ดํ„ฐ ๋“ฑ์— ๋งŽ์ด ์‚ฌ์šฉ ํ•œ ๋ฌธ์žฅ์— ํ•ด๋‹น ๋‹จ์–ด๊ฐ€ ์–ผ๋งˆ๋‚˜ ํฌํ•จ๋˜๊ณ  ์žˆ๋ƒ ๋“ฑ ์‹ค์Šต) # ๋นˆ์นธ ๊ธฐ์ค€์œผ๋กœ ๋ฌธ์ž์—ด ๋‚˜๋ˆ„๊ธฐ items = 'zero one two three'.split() print(items) # "," ๊ธฐ์ค€์œผ๋กœ ๋ฌธ์ž์—ด ๋‚˜๋ˆ„๊ธฐ example = 'python,jquery,javascript' print(example.split(",")) # ๋ฆฌ์ŠคํŠธ์˜ ๊ฐ ๊ฐ’์„ a, b, c, ๋ณ€์ˆ˜๋กœ unpacking example = 'python,jquery,javascript' a, b, c = example.split(",") # "." ๊ธฐ์ค€์œผ๋กœ ๋ฌธ์ž์—ด ๋‚˜๋ˆ„๊ณ  unpacking example = 'dami..
Pythonic Code "Life is short. Use Python" @Pythonic Code -๋‹ค๋ฅธ ์‚ฌ๋žŒ์˜ ์ฝ”๋“œ๋ฅผ ์ž˜ ์ดํ•ดํ•˜๊ธฐ ์œ„ํ•ด ํŒŒ์ด์ฌ ํŠน์œ ์˜ ๋ฌธ๋ฒ•์„ ํ™œ์šฉํ•ด ํšจ์œจ์ ์œผ๋กœ ์ฝ”๋“œ๋ฅผ ํ‘œํ˜„ํ•˜๋Š” ๊ธฐ๋ฒ• -๊ณ ๊ธ‰ ์ฝ”๋“œ๋ฅผ ์ž‘์„ฑ ํ•  ์ˆ˜๋ก ๋” ๋งŽ์ด ํ•„์š”ํ•จ @Why Pythonic Code? -๋งŽ์€ ๊ฐœ๋ฐœ์ž๋“ค์ด ํŒŒ์ด์ฌ ์Šคํƒ€์ผ๋กœ ์ฝ”๋”ฉํ•จ -๋‹จ์ˆœ for loop append๋ณด๋‹ค list๊ฐ€ ์ข€ ๋” ๋น ๋ฆ„ -์ต์ˆ™ํ•ด์ง€๋ฉด ์ฝ”๋“œ๋„ ์งง์•„์ง ์˜ˆ) ์—ฌ๋Ÿฌ ๋‹จ์–ด๋“ค์„ ํ•˜๋‚˜๋กœ ๋ถ™์ผ ๋•Œ # ์ผ๋ฐ˜ ์ฝ”๋“œ colors = ["a", "b", "c", "d", "e"] result = "" for s in colors: result += s print(result) # pythonic code colors = ["red", "blue", "green", "yellow"] res..
์‹ ๊ฒฝ๋ง ๋„คํŠธ์›Œํฌ ์ถœ๋ ฅ์˜ ๊ณ„์‚ฐ | Computing Neural Network Output ๋กœ์ง€์Šคํ‹ฑ ํšŒ๊ท€๋ฅผ ๋‚˜ํƒ€๋‚ด๋Š” ์ด ์›์€ ๋‘ ๋‹จ๊ณ„์˜ ๊ณ„์‚ฐ์„ ๋‚˜ํƒ€๋ƒ„ ์‹ ๊ฒฝ๋ง์—์„  ์ด๋ฅผ ๋งŽ์ด ๋ฐ˜๋ณตํ•จ ์€๋‹‰๋…ธ๋“œ ๋กœ์ง€์Šคํ‹ฑ ํšŒ๊ท€์™€ ๋น„์Šทํ•˜๊ฒŒ ์ด ๋…ธ๋“œ๋Š” ๋‘ ๋‹จ๊ณ„์˜ ๊ณ„์‚ฐ์„ ํ•จ a_i^[l]์—์„œ ๋Œ€๊ด„ํ˜ธ ์•ˆ์— ์žˆ๋Š” l์€ ์ธต ๋ฒˆํ˜ธ ์•„๋ž˜ ์ฒจ์ž i๋Š” ์ธต ์•ˆ์˜ ๋…ธ๋“œ ๋ฒˆํ˜ธ @๋ฒกํ„ฐํ™” ์กฐ์–ธ -ํ•œ ์ธต์— ๋…ธ๋“œ๊ฐ€ ์—ฌ๋Ÿฌ ๊ฐœ์ด๋ฉด ์„ธ๋กœ๋กœ ์Œ“๊ธฐ ์ด ์‹œ๊ทธ๋ชจ์ด๋“œ ํ•จ์ˆ˜๋Š” z^[1]์˜ ๊ฐ ์›์†Œ๋“ค์˜ ์‹œ๊ทธ๋ชจ์ด๋“œ ๊ฐ’์„ ๊ณ„์‚ฐํ•˜๋Š” ํ•จ์ˆ˜ ๋งˆ์ง€๋ง‰ ์ถœ๋ ฅ ์œ ๋‹›์„ ๋งค๊ฐœ ๋ณ€์ˆ˜ w์™€ b๋ฅผ ๊ฐ€์ง€๋Š” ๋กœ์ง€์Šคํ‹ฑ ํšŒ๊ท€์™€ ๋น„์Šทํ•˜๊ฒŒ ์ƒ๊ฐํ•˜๋ฉด, w๋Š” w^[2]T์™€ ๋น„์Šทํ•˜๊ณ  b๋Š” b^[2]์™€ ๋น„์Šทํ•จ ์€๋‹‰์ธต์ด ํ•˜๋‚˜์ธ ์‹ ๊ฒฝ๋ง์—์„œ๋Š” ์ถœ๋ ฅ๊ฐ’์„ ๊ณ„์‚ฐํ•˜๊ธฐ ์œ„ํ•ด ์œ„ ๋„ค ๋“ฑ์‹์„ ๊ตฌํ˜„ํ•˜๋ฉด ๋จ ์ด ์‹ ๊ฒฝ๋ง์˜ ์ถœ๋ ฅ๊ฐ’์„ ๊ณ„์‚ฐํ•˜๋Š” ๋ฐ์— ์ด ๋„ค ์ค„๋งŒ ํ•„์š” ์€๋‹‰์ธต์˜ ๋„ค ๋กœ์ง€์Šคํ‹ฑ ํšŒ๊ท€ ์œ ๋‹›์˜ ์ถœ๋ ฅ๊ฐ’ ๊ณ„์‚ฐ ์ถœ๋ ฅ์ธต์˜ ๋กœ์ง€์Šคํ‹ฑ ํšŒ๊ท€..
์‹ ๊ฒฝ๋ง ๋„คํŠธ์›Œํฌ ๊ตฌ์„ฑ | Neural Network Representations @์‹ ๊ฒฝ๋ง ๋„คํŠธ์›Œํฌ ๊ตฌ์„ฑ @์ž…๋ ฅ์ธต(Input layer) -์ž…๋ ฅ ํŠน์„ฑ๋“ค์˜ ์ธต -a^[0]์œผ๋กœ ํ‘œ๊ธฐ @์€๋‹‰์ธต(Hidden layer) -์ž…๋ ฅ์ธต๊ณผ ์ถœ๋ ฅ์ธต ์‚ฌ์ด์˜ ๋ชจ๋“  ์ธต -์€๋‹‰์ธต์ด๋ผ๋Š” ์ด๋ฆ„์€ ํ›ˆ๋ จ ์„ธํŠธ์—์„œ ๋ณผ ์ˆ˜ ์—†๋‹ค๋Š” ๊ฑธ ์˜๋ฏธ ์ง€๋„ ํ•™์Šต์œผ๋กœ ํ›ˆ๋ จ์‹œํ‚ค๋Š” ์‹ ๊ฒฝ๋ง์—์„  ํ›ˆ๋ จ ์„ธํŠธ๊ฐ€ ์ž…๋ ฅ๊ฐ’ X์™€ ์ถœ๋ ฅ๊ฐ’ Y๋กœ ์ด๋ฃจ์–ด์ ธ ์žˆ์Œ ์€๋‹‰์ธต์˜ ์‹ค์ œ ๊ฐ’์€ ํ›ˆ๋ จ ์„ธํŠธ์— ๊ธฐ๋ก๋˜์–ด ์žˆ์ง€ ์•Š์Œ ->ํ›ˆ๋ จ ์„ธํŠธ์—์„œ ๋ฌด์Šจ ๊ฐ’์ธ์ง€ ๋ชจ๋ฅธ๋‹ค๋Š” ๊ฒƒ @์ถœ๋ ฅ์ธต(Output layer) -์˜ˆ์ธก ๊ฐ’์ธ y hat์˜ ๊ณ„์‚ฐ์„ ์ฑ…์ž„์ง ์ž…๋ ฅ๊ฐ’๊ณผ ์ถœ๋ ฅ๊ฐ’์€ ์•Œ ์ˆ˜ ์žˆ์ง€๋งŒ ์€๋‹‰์ธต์˜ ๊ฐ’๋“ค์€ ์•Œ ์ˆ˜ ์—†์Œ @์ž…๋ ฅ์ธต ํ•˜๋Š” ์ผ a^[0] = X a^[0]์€ ์ž…๋ ฅ์ธต์˜ ํ™œ์„ฑ๊ฐ’์ด๋ผ๊ณ  ๋ถ€๋ฆ„ a๋Š” ํ™œ์„ฑ๊ฐ’์„ ์˜๋ฏธํ•˜๊ณ  ์‹ ๊ฒฝ๋ง์˜ ์ถฉ๋Œ์ด ๋‹ค์Œ ์ธต์œผ๋กœ ์ „๋‹ฌํ•ด์ฃผ๋Š” ๊ฐ’์„ ์˜๋ฏธ ์ž…๋ ฅ์ธต์€ X๋ฅผ ์€๋‹‰์ธต์œผ๋กœ..

Tiny Star