A web based to tool to analyse, understand and test Visuo-Textual Models.
- Visual Question Answering (VQA)
- Show Attend Ask and Answer (SAAA)
- Deep Modular Co-Attention Networks (MCANs)
- Visual Commonsense Reasoning (VCR)
- Recognition to Cognition (R2C)
- Custom question testing.
- Simultaneously test all models of a kind.
- User feedback on the model predictions.
- Human in the loop testing