Need help with pytorch-vqa?
Click the “chat” button below for chat support from the developer who created it, or find similar developers for support.

About the developer

143 Stars 69 Forks 10 Commits 5 Opened issues


Strong baseline for visual question answering

Services available


Need anything else?

Contributors list

Strong baseline for visual question answering

This is a re-implementation of Vahid Kazemi and Ali Elqursh's paper Show, Ask, Attend, and Answer: A Strong Baseline For Visual Question Answering in PyTorch.

The paper shows that with a relatively simple model, using only common building blocks in Deep Learning, you can get better accuracies than the majority of previously published work on the popular VQA v1 dataset.

This repository is intended to provide a straightforward implementation of the paper for other researchers to build on. The results closely match the reported results, as the majority of details should be exactly the same as the paper. (Thanks to the authors for answering my questions about some details!) This implementation seems to consistently converge to about 0.1% better results – there are two main implementation differences:

  • Instead of setting a limit on the maximum number of words per question and cutting off all words beyond this limit, this code uses per-example dynamic unrolling of the language model.
  • An issue with the official evaluation code makes some questions unanswerable. This code does not normalize machine-given answers, which avoids this problem. As the vast majority of questions are not affected by this issue, it's very unlikely that this will have any significant impact on accuracy.

A fully trained model (convergence shown below) is available for download.

Graph of convergence of implementation versus paper results

Note that the model in my other VQA repo performs better than the model implemented here.

Running the model

  • Clone this repository with:
    git clone --recursive
  • Set the paths to your downloaded questions, answers, and MS COCO images in
    • qa_path
      should contain the files
    • train_path
      should contain the train, validation, and test
      images respectively.
  • Pre-process images (93 GiB of free disk space required for f16 accuracy) with ResNet152 weights ported from Caffe and vocabularies for questions and answers with:
  • Train the model in
    This will alternate between one epoch of training on the train split and one epoch of validation on the validation split while printing the current training progress to stdout and saving logs in the
    directory. The logs contain the name of the model, training statistics, contents of
    , model weights, evaluation information (per-question answer and accuracy), and question and answer vocabularies.
  • During training (which takes a while), plot the training progress with:

Python 3 dependencies (tested on Python 3.6.2)

  • torch
  • torchvision
  • h5py
  • tqdm

We use cookies. If you continue to browse the site, you agree to the use of cookies. For more information on our use of cookies please see our Privacy Policy.