Skip to content
forked from MILVLG/openvqa

Implementation of the VQA model from my MSc project

License

Notifications You must be signed in to change notification settings

markvasin/openvqa

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Visual Question Answering (MSc Project)

This repository contains the implementation of the proposed VQA model from my MSc project. The implementation was done using PyTorch and OpenVQA framework.

VQA Model

The main folder for the model is located here, and the main python file for the model is here.

VQA model

Visualisation

The code to generate visualisation of attention distribution is here.

Visualisation of model's attention distribution on image

vis-1va

Visualisation of model's attention distribution on text data

vis-1va

Releases

No releases published

Packages

No packages published

Languages

  • Python 99.2%
  • Dockerfile 0.8%