• [NLP] ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators 정리

    안녕하세요. 이번에 제가 소개해드릴 내용은 현재 많은 자연어 처리 task에서 사용되고 있는 Pre-training Language Model (이하 PrLM) 중 하나인 ELECTRA입니다. 이 글은 아래 참고 자료 중에 중국어 자료을 바탕으로 정리했습니다. ELECTRA는 Efficiently Learning an Encoder that Classifies Token Replacements Accurately의 약자입니다. (앞선 BERT, ELMO에 이어 ELECTRA 약자도 발음하기도 편하고 고민한...


  • GuitarSolo (IT Convergence Guitar For Interactive Lesson)

    Research Team Chief of research : Soonuk Seol (suseol@koreatech.ac.kr) Cooperation researcher : Cho Gyeongmin, Lee Saehan, Lee Hojae, Choi Jongho, Kim Byeonggil (http://dame.co.kr/) Participating researcher : Shin Yejin, Lee Gangwon, Lee Jaeyeong, Park Jeonghyeok, Park Jinuk, Han Yumin Project Development Period 2015.11 ~ 2016.11 Explanation The GuitarSolo allows a user...


  • [Algorithm] GenomicRangeQuery

     [Codility] [PassingCars] [Code] # you can write to stdout for debugging purposes, e.g. # print("this is a debug message") def solution(A): # write your code in Python 3.6 answer = 0 zero_cnt = 0 one_cnt = 1 for i in A: if i == 0: zero_cnt +=1 if i...


  • [Algorithm] MinAvgTwoSlice

     [Codility] [MinAvgTwoSlice] [Code] # you can write to stdout for debugging purposes, e.g. # print("this is a debug message") def solution(A): # # write your code in Python 3.6 # accum_A = [0] # for i in range(len(A)): # accum_A.append(accum_A[i]+A[i]) # # print(accum_A) # min_num = (A[0]+A[1]) /2 #...


  • [Algorithm] CountDiv

     [Codility] [CountDiv] [Code] # you can write to stdout for debugging purposes, e.g. # print("this is a debug message") def solution(A, B, K): # write your code in Python 3.6 # print(A, B, K) if A %K != 0: if A > K: A += A %K else: A...