In this repo, I have written and refactored some code to mirror strings using the seq2seq model. For example, if the input is ABC, the ground-truth output is CBA.
- https://pytorch.org/tutorials/intermediate/seq2seq_translation_tutorial.html#sphx-glr-intermediate-seq2seq-translation-tutorial-py
- https://arxiv.org/abs/1409.3215
- https://github.com/vanhuyz/udacity-dl/blob/master/6_lstm-Problem3.ipynb
CUDA 9.0 + Pytorch 1.0
ROOT = path/to/this/repo
- download the dataset
cd ROOT
python dataset/download.py
- training:
python main.py
- inference:
python inference.py
attention, word embedding (glove or bert)