MBART_50_STREAMLIT_TRANSLATOR

• AN OCR AND MULTILINGUAL MACHINE TRANSLATOR USING 🤗HUGGING FACE

• Found an image with unknown language?? Use my application to extract text from it and translate.
• Upload a high quality image with proper font and get the required TRANSLATIONS.

REQUIREMENTS:

• python 3.7+
• pytorch == 1.9.0
• streamlit == 0.86.0

CREATING VENV IN CONDA:

conda create --name myenv

Using a virtual environment:

conda activate myenv

Deactivating a virtual environment:

conda deactivate myenv

INSTALLING DEPENDENCIES:

PYTORCH FOR CUDA 10.2:

conda install pytorch torchvision torchaudio cudatoolkit=10.2 -c pytorch

PYTORCH - CPU ONLY VERSION:

conda install pytorch torchvision torchaudio cpuonly -c pytorch

HUGGING FACE TRANSFORMERS 🤗:

pip install transformers

STREAMLIT:

pip install streamlit == 0.86.0

RUNNING THE APPLICATION:

run :

streamlit run app.py

• Jump to http://localhost:8501
• Your app will be hosted in the local server.
• Then deploy the app to cloud platforms but in our case, the free tier versions of Heroku and Streamlit Sharing won't handle whole 2.4GB pre-trained model (MBART-50).
• The home and the output pages of the web app is shown below: