Shiny new things: I've already updated this solution for 2021, this include:
- New word embedding analysis in assignment 1
- New assignment 5 on pretraining and synthesizer transformer
- Simplified NMT assignment 4
My solutions for CS224N 2019
- Course page CS224N
- Lecture videos 2019 Youtube
- Stanford online hub
- Install Chrome extention Text all the thing for better formula display
- Since Kaggle updated their GPUs into P100 (2 x performance compare to K80), this Kaggle kernel will show how I trained A5 using Kaggle kernel (around 6 hours)