Optimizing transformer-based machine translation model for single GPU training: a hyperparameter ablation study

Open in new window