Kolle-D/Oku-Mini-Model

A simple model to translate oku language, an indigenous language found in the northern part of cameroon. 1. Clone the repository ↓ 2. Install the requirements ↓ 3. Run preprocessing.py ↓ 4. Run train.py ↓ 5. A new models/ folder is created automatically ↓ 6. Run inference.py or the Streamlit app 8. For example the steps above are outlined below with their commands 9. git clone github.com cd OkuTranslator pip install -r requirements.txt python src/preprocessing.py python src/train.py After Training finishes you should have this OkuTranslator/ │ ├── models/ │ └── oku_model/ │ ├── config.json │ ├── model.safetensors │ ├── tokenizer.json │ ├── tokenizer_config.json │ ├── special_tokens_map.json │ └── generation_config.json python src/inference.py streamlit run app/streamlit_app.py