Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/navaneethelite/mbart_50_streamlit_translator
AN OCR AND MULTILINGUAL MACHINE TRANSLATOR USING 🤗HUGGING FACE. Upload a high quality image with proper font and get the required TRANSLATIONS.
https://github.com/navaneethelite/mbart_50_streamlit_translator
huggingface nlp pytorch streamlit
Last synced: 2 months ago
JSON representation
AN OCR AND MULTILINGUAL MACHINE TRANSLATOR USING 🤗HUGGING FACE. Upload a high quality image with proper font and get the required TRANSLATIONS.
- Host: GitHub
- URL: https://github.com/navaneethelite/mbart_50_streamlit_translator
- Owner: NAVANEETHELITE
- License: mit
- Created: 2021-08-29T07:14:53.000Z (over 3 years ago)
- Default Branch: master
- Last Pushed: 2022-04-18T13:50:33.000Z (almost 3 years ago)
- Last Synced: 2024-06-09T15:25:07.026Z (8 months ago)
- Topics: huggingface, nlp, pytorch, streamlit
- Language: Python
- Homepage: https://github.com/NAVANEETHELITE/MBART_50_STREAMLIT_TRANSLATOR
- Size: 621 KB
- Stars: 4
- Watchers: 1
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# 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:
```sh
conda create --name myenv
```
Using a virtual environment:
```sh
conda activate myenv
```
Deactivating a virtual environment:
```sh
conda deactivate myenv
```
### INSTALLING DEPENDENCIES:
PYTORCH FOR CUDA 10.2:
```sh
conda install pytorch torchvision torchaudio cudatoolkit=10.2 -c pytorch
```
PYTORCH - CPU ONLY VERSION:
```sh
conda install pytorch torchvision torchaudio cpuonly -c pytorch
```
HUGGING FACE TRANSFORMERS 🤗:
```sh
pip install transformers
```
STREAMLIT:
```sh
pip install streamlit == 0.86.0
```
### RUNNING THE APPLICATION:
run :
```sh
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:
