Demos
Classifying PDF documents using text-based classification models is a powerful capability Generative AI Lab provides. Users can now pre-annotate and classify images and PDF documents with over 1500 pre-trained models available in the shared repository – Models Hub.
Manual classification is also supported, allowing users to classify documents themselves. This enables document classification in its original form, preserving the integrity of PDFs without converting them to plain text.
New Project Type- Visual NLP Classification:
Using this feature is easy: select the project type called “Visual NLP Classification” from the exiting project template collection of Generative AI Lab and start configuring it. Here are the steps:
- Go to the “Content Type” page.
- Select the “Image” tab.
- Choose “Visual NLP Classification” as the project type.
Visual Classification with Generative AI Lab
Users can classify both images and PDFs using their original form. This means working with the complete/original document, preserving its layout and content, rather than just classifying extracted text. Classification is easy and the workflow and user interface are consistent with previous implementations:
1. Pre-annotation using Classification Models:
- After selecting the project type, go to the “Reuse Resource” page.
- Choose a classification model from the available pre-trained models.
- Save the configuration.
- Import OCR Documents
- Once the tasks are imported, click on the pre-annotate button to classify tasks based on classification models.
2. Manual Classification:
- After selecting the project type, go to the “Customize Labels” page.
- Click on the Choices tab and Add/Remove choices for classification
- Click on “Code” view and change the choice property in Choice tag to multiple to enable multiple classification.
- Save the configuration.
- Import OCR Documents
- Open the tasks and classify them manually.
Getting Started is Easy
Generative AI Lab is a text annotation tool that can be deployed in a couple of clicks using either Amazon or Azure cloud providers, or installed on-premise with a one-line Kubernetes script.
Get started here: https://nlp.johnsnowlabs.com/docs/en/alab/install
