Token prediction using deep learning
This project is part of my master curriculum at Technical University of Darmstadt, Germany. This project aims to predict missing tokens in a javascript file. The project has been implemented using tensorflow. 800 javascript file has been used to train the model and another 200 to test it. The accuracy was more than 80% while the hole size(number of consecutive missing token) was 1 and more than 35% while the hole size was 3. Details in the 'project description' and 'project report' of the project.