CameroonFranceChapter_LegalComplianceAssistant

Models

Gliner is used for entity extraction on text based data

YOLO model is used for object detection on Signature and Logo

Steps to run the application

1. First inside the src/model_inferencing directory run

   1.1) git clone https://github.com/ultralytics/yolov5

   1.2) create a new folder runs/train/exp/weights and paste the weights best.pt and last.pt

   1.3) Drive link of YOLO fine tuned model: https://drive.google.com/drive/folders/1HYh70f5pWsUXggeNHLMTCU8mTP4ccTpc?usp=drive_link

2. Create a new folder inside src/model_inferencing directory called gliner_inference_model and paste the contents from drive link into this directory

   Drive link: https://drive.google.com/file/d/1e2sKYOgQS3a1LM-xuT0JL2EzQJfGDv7K/view?usp=drive_link

3. Create a new output folder in root directory

4. Install Docker desktop in your machine

5. Finally in the root folder, run the following command docker-compose up --build

6. The streamlit app runs in localhost:8501 and the user can upload the file

7. The backend api runs which does model inferencing runs in localhost:8000 port

8. To shutdown the , docker-compose down