Document Classification Models

🤝 Community Help
1
  • Project Type: Work Project
  • Operating System & Browser: Windows 11 & MS edge
  • Project Universe Link or Workspace/Project ID: None. This isn’t on Roboflow.

I am searching for any good models that can perform document Classification. My documents are PDFs that are structured. They are invoices, delivery notes, identity documents etc.

I am building a classifier that can run in CPU which will perform classification of these documents based on the content in the docs. I will be converting the pdfs and then passing to an image classifier.

I need help on finding which lightweight classifiers are best for the task. Data is enough for training. I need to know of a good classifier for this purpose.

This is my first post so I apologise for any things missing or not following the formats properly. Been using roboflow for quite some time and knew about it from Piotr on LinkedIn.

2

If you’re looking to train your own document classifier, I’d recommend using ResNet for a small/fast option to run on CPU. A ViT would be more accurate though may not run on your CPU. Upload your documents, assign them class labels, and train a model. Both are able to be custom trained in Roboflow today.

1 Like
3

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.