https://pytorch.org/tutorials/intermediate/seq2seq_translation_tutorial.html NLP From Scratch: Translation with a Sequence to Sequence Network and Attention — PyTorch Tutorials 1.4.0 documentation Note Click here to download the full example code NLP From Scratch: Translation with a Sequence to Sequence Network and Attention Author: Sean Robertson This is the third and final tutorial on doing “N..

출처 1) Neural Machine Translation By Jointly Learning to Align and Translate 2) Attention: Illustrated Attention 3) Attention and Memory in Deep Learning and NLP 기존 Encoder-Decoder RNN/LSTM 모델의 문제점 - 아무리 긴 input sentence가 주어져도 고정 길이 벡터fixed-length vector로 압축해서 표현해야 함 - Decoder는 Encoder의 마지막 은닉상태만 전달받음 → 엄청 긴 문장이라면 엄청 많이 까먹음 기존 Encoder-Decoder RNN/LSTM 모델의 문제점 해결 - 고정길이벡터 X - input sentence는 여러 벡터..

https://www.deeplearningwizard.com/deep_learning/practical_pytorch/pytorch_recurrent_neuralnetwork/ Recurrent Neural Networks (RNN) - Deep Learning Wizard Recurrent Neural Network with PyTorch About Recurrent Neural Network Feedforward Neural Networks Transition to 1 Layer Recurrent Neural Networks (RNN) RNN is essentially an FNN but with a hidden layer (non-linear output) that passes on informa..

https://towardsdatascience.com/illustrated-guide-to-recurrent-neural-networks-79e5eb8049c9 Illustrated Guide to Recurrent Neural Networks Understanding the Intuition towardsdatascience.com RNN과 은닉상태 초기화 → input 단어와 은닉상태를 RNN에 집어 넣기 → 그 단어의 output과 새로운 은닉상태가 출력됨 → 이 출력을 다시 RNN에 집어넣기 → 단어 없을 때까지 반복 → 출력을 FeedForward 레이어에 넣기 → 최종결과(prediction) 출력 은닉상태가 RNN의 기억장치 (저번 출력을 다음 입력으로 넘겨줌) tanh 함수는 출력을 [-..

https://medium.com/@jongdae.lim/%EA%B8%B0%EA%B3%84-%ED%95%99%EC%8A%B5-machine-learning-%EC%9D%80-%EC%A6%90%EA%B2%81%EB%8B%A4-part-1-9a0297198ad8 기계 학습(Machine Learning, 머신 러닝)은 즐겁다! Part 1 (원문 : Machine Learning is Fun! by Adam Geitgey) medium.com 맞았는지/틀렸는지 여부 + 나의 대답 True Positives : O라고 답했고 맞음 (정답O, 나의 출력O) True Negatives : X라고 답했고 맞음 (정답X, 나의 출력X) False Positives : O라고 답했고 틀림 (정답X, 나의 출력O) Fa..