There are 4 folders
Neural Machine Translation Folder: Includes nmt.ipnyb. This file contains Machine Learning part of the project. It contains code for input preprocessing, composing of neural network model, training the model, translate inputs and saving model. The model was created based on the neural machine translation guide prepared by Tensorflow. Guide can be found here: https://www.tensorflow.org/text/tutorials/nmt_with_attention
Presentation Folder: This folder includes my project presentation.
User Interface Folder: The user interface of language translation tool. Flask and HTML were used for this part. Also this folder includes all 14 language translation models.
Dataset Folder: Includes necessary files for creating datasets. Copy of Form Templates of All Languages.xlsx contains English form links and their counterparts from 14 different languages. Main.py is a web crawler. It crawls the forms of excel file and extracts their form ids'. DatasetGenerator.ipnyb uses these ids and also Jotform API, extracts the texts of forms and creates datasets from these texts. Then it merges these datasets with datasets available at http://www.manythings.org/anki/ and by merging them it prepares a final dataset for each 14 languages. datasets folder contains these datasets and formid folder includes ids'of forms