Yet another sequence to sequence model with multiple attention mechanisms in Pytorch. This code is heavily inspired by the official torch tutorial, with some differences, most importantly the batched training and inference functionality.
To use the code you need python > 3.5. Then install the required packages using pip:
pip install -r requirements.txt
For training, we use train.py file which can be invoked with additional flags e.g.:
python3 train.py -lr 0.01 -h 256 -d 'cuda' -e 150