The scripts for extracting from the epub version of the Jehovah Witness Bible and the online Igbo English Dictionary can be found in the data-tools folder.
The seq2seq, rnn-attention models can be found in the models folder.
The transformer model can be found in the transformer folder.
The jupyter notebooks can be found in the notebooks folder.
The online dictionary dataset can be found here
The Jehovah Witness Bible dataset can be found here.
The design choices I made for this project along with more information on the Machine Translation models can be found in this blog post I wrote here