linjieli222 / VQA_ReGAT

Research Code for ICCV 2019 paper "Relation-aware Graph Attention Network for Visual Question Answering"

https://arxiv.org/abs/1903.12314

linjieli222/VQA_ReGAT Issues

unhandled cuda error
Updated a month ago1
Can you provide a well-trained model?
Updated a year ago1
Loss can't backward
Closed 3 years ago2
download.sh
Closed 4 years ago3
attention map
Updated 2 years ago
Can you tell me the specific label of the semantic relationship type?
Updated 3 years ago
NOthing
Closed 3 years ago2
features/model that are interrupted during download doesn't continue from the last checkpoint
Updated 3 years ago5
How to test your model with image and text input by a user?
Closed 3 years ago2
Can you provide a Google Drive link for dataset?
Closed 4 years ago2
The semantic_embedding and spatic_embedding types.
Updated 3 years ago4
weighted sum confusion
Updated 3 years ago2
What are the specific categories in your semantic relationship classification?
Closed 4 years ago1
pos_box/bb
Closed 3 years ago2
about weighted sum of the three modules
Closed 3 years ago
Low accuracy re-train with the provided config files of the pretrained models
Closed 4 years ago12
Wdir(i,j) in Function 8 in the explicit model
Closed 3 years ago6
Questions for categories
Closed 3 years ago2
How can I compute sem_adj_matrix, spa_adj_matrix ？
Closed 4 years ago1
Question about dir(i,j) matrix in equation(8)
Closed 4 years ago2
Learning rate related issues
Closed 3 years ago1
Does the setting of num_workers in DataLoder affect the final result？
Closed 3 years ago1
Some question about semantic relationship classification
Closed 3 years ago1
Load the cake/val_target. pkl
Closed 4 years ago1
Some questions about the union bounding box feature vector for classifer mode
Closed 4 years ago1
Training with Visual Genome
Closed 4 years ago4
Code for learned image and question attentions
Closed 4 years ago1
loading features from the fixed h5 file
Closed 4 years ago1
Training confusion
Closed 4 years ago1
Code for extracting image features
Closed 4 years ago1
GPU will low memory
Closed 5 years ago2
There may be a better memory allocation scheme when loading datasets
Closed 5 years ago3
Can you update a new download.sh?
Closed 4 years ago
about spa_adj_matrix & sem_adj_matrix
Closed 4 years ago1
GAT Loss Function
Closed 4 years ago1
The difference between implicit relation and spatial relation.
Closed 4 years ago1
Could you please share the code on how to get semantic and spatial information in the .hdf5 files?
Closed 4 years ago2
Where do you add b_{lab(i,j)} to your value term (see eq{2} in the paper)
Closed 4 years ago1
Code to build the spatial adjacency matrix
Closed 4 years ago3
Best parameter configuration？
Closed 5 years ago1
Test with new images
Closed 5 years ago1