This project aims at implementing the Neural Machine Translation model to automate the translation of Hindi sentences into English sentences. The models are implemented using the PyTorch Framework.
This project was completed at the IITK in the course CS779-Statistical NLP. The whole competition was distributed in the 4 phases, where in each phase we need to submit the predictions against the test-dataset released for that phase.
All datasets (train and test-phase datasets) are present in the 'Datasets' directory.
For the competition, various techniques were implemented. Some of them include Seq-2-Seq model, Seq-2-Seq with attention mechanism, Transformers etc.
All the models for different phases are present in the directory.