Printed / Written scanned text document classification

The training took place on Google Colab using RVL-CDIP-small dataset with 200 examples per class on invoice as printed and handwritten classes.

Set up

python3.8 -m venv venv 
source venv/bin/activate
pip install -r requiements.txt

Inference

Change the image_path variable in the inference.py. Run with:

python3 inference.py

The assertion at the end is added to verify the results.

References

• Original repo taken as the base
• Dataset