zikunukiz / NMT-MCTS

Neural Machine Translation with Monte-Carlo Tree Search

Code for MCTS is in folder 'MultiProcess' and code for replicating baseliens is in 'BaselineCode' We already ran the tokenizer from 'fairseq' on the IWSLT14 dataset and have the train,validation, and test set files in IWSLTTokenizedData

Course Website: https://duvenaud.github.io/learning-to-search/