CrispenGari/nmt

🎗 Neural Machine Translation (nmt) repository contains an end-to-end deep artificial neural network for machine translation and language identification based on south-african-languages. ### Neural Machine Translation (nmt) and Language Identification (li). 🎗 This repository contains an end-to-end deep artificial neural network for machine translation and language identification based on south-african-languages. ### Notebooks All the notebooks that were used in this project are found in the `notebooks` folder. Here is the folder structure: ```shell notebooks: ├───bi-nmt ├───lang-identification │ ├───datasets │ └───model ├───languages └───nmt-gathering ``` ### Language Identification In this model we created a language identification model that identifies `6` south african languages. These languages are as follows: ``` ``` The model was able to perform very well during inference and we evaluated with the following metrics on the test examples. 1. confusion matrix 2. classification report ```shell precision recall f1-score support af 1.00 0.99 0.99 4542 ts 1.00 0.98 0.99 4604 st 0.98 0.99 0.98 4462 en 0.96 1.00 0.98 4692 xh 0.98 0.97 0.98 4571 zu 0.98 0.97 0.98 4651 accuracy 0.98 27522 macro avg 0.98 0.98 0.98 27522 weighted avg 0.98 0.98 0.98 27522 ``` 3. loss The model was trained and be able to have a loss reduced to `0.058` on test examples. 4. accuracy The model was trained and be able to obtain an accuracy of `98.35% ` on test examples. ### License In this repository I'm using the `MIT` License which reads as follows: ``` MIT License Copyright (c) 2022 crispengari Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to wh …