yashkant / concat-vqa Goto Github PK
View Code? Open in Web Editor NEWOfficial code for the paper "Contrast and Classify: Training Robust VQA Models" published at ICCV, 2021
Home Page: https://yashkant.github.io/projects/concat-vqa.html
Official code for the paper "Contrast and Classify: Training Robust VQA Models" published at ICCV, 2021
Home Page: https://yashkant.github.io/projects/concat-vqa.html
Hello,
For the table 2 in the paper, where CS score is reported, is it CS or CS-BT in the code? CS is computed for revqa, and CS-BT is computed for revqa_bt.
Hi, I want to download the data to run the model.
However, I can't download the files from dropbox. (the website https://www.dropbox.com/sh/v826l3ge7oz4vz4/AACRimDdy_BGnN2XZJDZLyY6a?dl=0)
Hi,
Thank you for sharing the code. I am wondering if you failed to upload "val_target.pkl" to the split folder?
Kind regards,
Hello! Thank you for your great work. I was able to download files from drop box, however, could you add the description of data sets in split directory? Because I am not sure which files correspond to VQA V2, which to VQA-paraphrase and so on. E.g. splits/questions_train_aug.pkl.
Jurijs
Hello,
Thanks for the code and provided extracted features for images. However, I would like to see corresponding images, where should I get them? I see that data contains information about image id, however, I am not sure where are these images.
Thanks,
Jurijs
Dear scholar,
extract_features.py didn't have the params named "imdb_gt_file" , so how to fulfill the object "For grounding truth file to extract features "
python data-release/extract_features.py --model_file data/detectron_model.pth --config_file data/detectron_config.yaml --imdb_gt_file <path_to_imdb_npy_file_generated_above> --output_folder <path_to_output_extracted_features>
Dear scholar,
I found the link vqa-maskrcnn-benchmark didn't have the folder named as "data-release",where I can download it? So I can't run the instructions about the below procedures
cd data-release
wget https://dl.fbaipublicfiles.com/vilbert-multi-task/detectron_model.pth
wget https://dl.fbaipublicfiles.com/vilbert-multi-task/detectron_config.yaml
To extract features for images, run from root directory:
python data-release/extract_features.py --model_file data/detectron_model.pth --config_file data/detectron_config.yaml --image_dir <pa
Dear scholar,
in data-release/extract_features.py, model.to("cuda") will throw errors. and the code is broken. Could you have some suggestion?
def _build_detection_model(self):
cfg.merge_from_file(self.args.config_file)
#cfg.merge_from_list(["MODEL.DEVICE", "cuda"])
cfg.freeze()
model = build_detection_model(cfg)
checkpoint = torch.load(self.args.model_file, map_location=torch.device("cpu"))
load_state_dict(model, checkpoint.pop("model"))
model.to("cuda")
model.eval()
return model
Hello,
Could you please share the script to generate a list of negatives? Negatives, which we load from .pkl?
Jurijs
Dear scholar,
extract_features.py didn't have the params named "imdb_gt_file" , so how to fulfill the object "For grounding truth file to extract features "
python data-release/extract_features.py --model_file data/detectron_model.pth --config_file data/detectron_config.yaml --imdb_gt_file <path_to_imdb_npy_file_generated_above> --output_folder <path_to_output_extracted_features>
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.