CNN ํ…์ŠคํŠธ ์œ ์‚ฌ๋„ ๋ถ„์„(Feat. Quora pairs)
ยท
๐Ÿ—ฃ๏ธ Natural Language Processing
kaggle www.kaggle.com/c/quora-question-pairs/submissions Quora ์งˆ๋ฌธ ๋‹ต๋ณ€ ์‚ฌ์ดํŠธ์—์„œ ๊ฐ™์€ ์งˆ๋ฌธ์— ๋Œ€ํ•œ ํŒ๋ณ„ ๋ฌธ์ œ column # ['id', 'qid1', 'qid2', 'question1', 'question2', 'is_duplicate'] Quora Question Pairs Can you identify question pairs that have the same intent? www.kaggle.com CNN - ํ•ฉ์„ฑ ์‹ ๊ฒฝ๋ง In deep learning, a convolutional neural network (CNN, or ConvNet) is a class of deep neural networks, most commonly applied ..
XG ๋ถ€์ŠคํŠธ(eXtream Gradient Boosting)
ยท
๐Ÿ‘พ Deep Learning
์•™์ƒ๋ธ” ๋ชจ๋ธ ์ค‘ ํ•˜๋‚˜์ธ XG ๋ถ€์ŠคํŠธ(eXtream Gradient Boosting)๋Š” ์บ๊ธ€ ์‚ฌ์šฉ์ž์—๊ฒŒ ํฐ ์ธ๊ธฐ๋ฅผ ์–ป๊ณ  ์žˆ๋Š” ๋ชจ๋ธ์ด๋‹ค. *์•™์ƒ๋ธ”: ์—ฌ๋Ÿฌ ๊ฐœ์˜ ํ•™์Šต ์•Œ๊ณ ๋ฆฌ์ฆ˜์„ ์‚ฌ์šฉํ•ด ๋” ์ข‹์€ ์„ฑ๋Šฅ์„ ์–ป๋Š” ๋ฐฉ๋ฒ• ์•™์ƒ๋ธ”์—๋Š” ๋ฐฐ๊น…๊ณผ ๋ถ€์ŠคํŒ…์ด ์žˆ๋‹ค. ensemble ์ข…๋ฅ˜ : single(CNN,RNN) bagging boosting *๋ฐฐ๊น…: ์—ฌ๋Ÿฌ ๊ฐœ์˜ ํ•™์Šต ์•Œ๊ณ ๋ฆฌ์ฆ˜, ๋ชจ๋ธ์„ ํ†ตํ•ด ๊ฐ๊ฐ ๊ฒฐ๊ณผ๋ฅผ ์˜ˆ์ธกํ•˜๊ณ  ๋ชจ๋“  ๊ฒฐ๊ณผ๋ฅผ ๋™๋“ฑํ•˜๊ฒŒ ๋ณด๊ณ  ์ทจํ•ฉํ•ด์„œ ๊ฒฐ๊ณผ๋ฅผ ์–ป๋Š” ๋ฐฉ์‹ *๋ถ€์ŠคํŒ…: ๋ฐฐ๊น…๊ณผ ๋‹ค๋ฅด๊ฒŒ ๋ชจ๋ธ์˜ ๊ฒฐ๊ณผ๋ฅผ ์ˆœ์ฐจ์ ์œผ๋กœ ์ทจํ•ฉ, ๋‹จ์ˆœํžˆ ํ•˜๋‚˜์”ฉ ์ทจํ•ฉํ•˜๋Š” ๋ฐฉ๋ฒ•์ด ์•„๋‹ˆ๋ผ ์ด์ „ ์•Œ๊ณ ๋ฆฌ์ฆ˜, ๋ชจ๋ธ์ด ํ•™์Šต ํ›„ ์ž˜๋ชป ์˜ˆ์ธกํ•œ ๋ถ€๋ถ„์— ๊ฐ€์ค‘์น˜๋ฅผ ์ค˜์„œ ๋‹ค์‹œ ๋ชจ๋ธ๋กœ ๊ฐ€์„œ ํ•™์Šตํ•˜๋Š” ๋ฐฉ์‹ XG ๋ถ€์ŠคํŠธ๋Š” ๋ถ€์ŠคํŒ… ๊ธฐ๋ฒ• ์ค‘ ํŠธ๋ฆฌ๋ถ€์ŠคํŒ…(Tree Boosting) ๊ธฐ๋ฒ•..
KoNLPy ์ข…๋ฅ˜
ยท
๐Ÿ—ฃ๏ธ Natural Language Processing
# KoNLPy ๋ผ์ด๋ธŒ๋Ÿฌ๋ฆฌ ๊ฐ์ฒด # ์ฃผ๋กœ Okt ๋ถ„์„๊ธฐ๋ฅผ ์‚ฌ์šฉํ•จ # Hannanum: ํ•œ๋‚˜๋ˆ”. KAIST Semantic Web Research Center ๊ฐœ๋ฐœ. # http://semanticweb.kaist.ac.kr/hannanum/ # Kkma: ๊ผฌ๊ผฌ๋งˆ. ์„œ์šธ๋Œ€ํ•™๊ต IDS(Intelligent Data Systems) ์—ฐ๊ตฌ์‹ค ๊ฐœ๋ฐœ. # http://kkma.snu.ac.kr/ # Komoran: ์ฝ”๋ชจ๋ž€. Shineware์—์„œ ๊ฐœ๋ฐœ. # https://github.com/shin285/KOMORAN # Mecab: ๋ฉ”์นด๋ธŒ. ์ผ๋ณธ์–ด์šฉ ํ˜•ํƒœ์†Œ ๋ถ„์„๊ธฐ๋ฅผ ํ•œ๊ตญ์–ด๋ฅผ ์‚ฌ์šฉํ•  ์ˆ˜ ์žˆ๋„๋ก ์ˆ˜์ •. # https://bitbucket.org/eunjeon/mecab-ko # Open Korean Text: ์˜คํ”ˆ ์†Œ..
SVD(singular value decomposition) VS SVM(support vector machine)
ยท
๐Ÿ—ฃ๏ธ Natural Language Processing
dspace.mit.edu/bitstream/handle/1721.1/77902/18-337j-spring-2005/contents/lecture-notes/chapter_12.pdf SVM : categorical data classifier SVD : Component decomposition
[Ubuntu] Vim plugin ์„ค์น˜(๋“ค์—ฌ์“ฐ๊ธฐ, ์ž๋™๊ด„ํ˜ธ)
ยท
๐Ÿ› ๏ธ Tools
vim์„ ํ”Œ๋Ÿฌ๊ทธ์ธ ์ด๋‚˜ ๋ณ„๋‹ค๋ฅธ ์„ค์ • ์—†์ด ์‚ฌ์šฉํ•  ๊ฒฝ์šฐ ์ด๊ฑธ ์™œ ์“ธ๊นŒ? ์‹ถ์„ ์ •๋„์˜ ๋ถˆํŽธํ•œ ๋ถ€๋ถ„์ด ๋งŽ๋‹ค. ํ•˜์ง€๋งŒ Plugin์ด๋‚˜ ์„ค์ •์„ ํ†ตํ•ด ์ž์‹ ์˜ ์ž…๋ง›์— ๋งž๋Š” ๊ฐœ๋ฐœํ™˜๊ฒฝ์„ ๊พธ๋ฐ€ ์ˆ˜ ์žˆ๋‹ค. ๋‚˜๋Š” ์ž๋™ ๊ด„ํ˜ธ์— ๋„ˆ๋ฌด ์ต์ˆ™ํ•ด์„œ ์ž๋™ ๊ด„ํ˜ธ ํ”Œ๋Ÿฌ๊ทธ์ธ ์„ค์น˜๋กœ ์˜ˆ๋ฅผ ๋“ค๊ฒ ๋‹ค. ์ž๋™ ๊ด„ํ˜ธ ํ”Œ๋Ÿฌ๊ทธ์ธ์ธ delimitMate๋ฅผ ์ด์šฉํ•˜๋ฉด jetbrain IDE์ฒ˜๋Ÿผ ์ž๋™๊ด„ํ˜ธ๋ฅผ ๋งŒ๋“ค์ˆ˜์žˆ๋‹ค. ๋จผ์ € ์—…๋ฐ์ดํŠธ๋ฅผ ํ•˜์ž $sudo apt-get update ์„ค์น˜ $sudo apt-get install vim ๊ธฐ๋Šฅ ์ถ”๊ฐ€ $vi ~/.vimrc set number " line ํ‘œ์‹œ set ai " auto indent set si " smart indent set cindent " c style indent set shiftwidth=4 " ์ž๋™..
[Ubuntu] ์šฐ๋ถ„ํˆฌ ์„œ๋ฒ„์— ์ž๋ฐ” ์„ค์น˜
ยท
๐Ÿ› ๏ธ Tools
https://all-record.tistory.com/181 [Ubuntu] ์šฐ๋ถ„ํˆฌ ์„œ๋ฒ„(16.04)์— ์ž๋ฐ” ์„ค์น˜ ์šฐ๋ถ„ํˆฌ๋ฅผ ์„œ๋ฒ„์— ์ž๋ฐ”๋ฅผ ์„ค์น˜ํ•ด ๋ณด์ž. ์—ฌ๊ธฐ์—์„œ๋Š” openjdk-8์„ ์„ค์น˜ํ•  ๊ฒƒ์ด๋‹ค. ์šฐ๋ถ„ํˆฌ ์„œ๋ฒ„์— ์ž๋ฐ” ์„ค์น˜ JDK์™€ JRE ์„ค์น˜ ๋ช…๋ น์–ด๋ฅผ ์‹คํ–‰ํ•œ๋‹ค. # JRE, JDK ์„ค์น˜ sudo apt-get install openjdk-8-jre sudo apt-get i.. all-record.tistory.com ์ฐธ๊ณ  https://elfinlas.tistory.com/365 Ubuntu 16.04 LTS์—์„œ Oracle Java 8 ์„ค์น˜ํ•˜๊ธฐ ์ด๋ฒˆ์— ์ง‘์—์„œ ์‚ฌ์šฉํ•˜๋˜ ๋…ธํŠธ๋ถ ์„œ๋ฒ„๋ฅผ ๋ณ€๊ฒฝํ•˜๋ฉด์„œ ๊ธฐ์กด์˜ Ubuntu 12.04 LTS์—์„œ Ubuntu 16.04 LTS๋กœ ์ƒˆ๋กœ ์„ค์น˜ํ•˜์˜€๋‹ค. ์„ค์น˜ํ•˜๋ฉด์„œ Ja..
๋‹คํ–ˆ๋‹ค