Open yoonhoelee opened 3 years ago
키워드: RNN, LSTM, GRU, Transform
어려웠던 부분: LSTM과 GRU를 하는 이유는 이해를 했다. 하지만 두 모델이 학습하는 방안을 완벽하게 이해를 했다고 보기는 힘들 것 같다.
더 알고 싶은 부분 / 공부하고 싶은 부분: Papers with code에서 tacotron2를 사용한 모델을 가져와서 구현해봤으나, 코드를 완벽하게 이해하지 못했다. 이부분에 대해서 더 많은 공부를 해봐야 될것 같다.
느낀 점: Papers with code에 나온 논문을 구현하는 작업을 봤는데, 언젠가 나도 이런 것을 스스로 구현할 수 있을까에 대한 불신이 생긴다.
키워드: RNN, LSTM, GRU, Transform
어려웠던 부분: LSTM과 GRU를 하는 이유는 이해를 했다. 하지만 두 모델이 학습하는 방안을 완벽하게 이해를 했다고 보기는 힘들 것 같다.
더 알고 싶은 부분 / 공부하고 싶은 부분: Papers with code에서 tacotron2를 사용한 모델을 가져와서 구현해봤으나, 코드를 완벽하게 이해하지 못했다. 이부분에 대해서 더 많은 공부를 해봐야 될것 같다.
느낀 점: Papers with code에 나온 논문을 구현하는 작업을 봤는데, 언젠가 나도 이런 것을 스스로 구현할 수 있을까에 대한 불신이 생긴다.