batra-mlp-lab / visdial-challenge-starter-pytorch

Starter code in PyTorch for the Visual Dialog challenge

https://visualdialog.org/challenge/2019

batra-mlp-lab/visdial-challenge-starter-pytorch Issues

why do max_sequence_length - 1 in dataset.py
Closed 2 years ago
resource of the Faster-RCNN pre-trained on Visual Genome
Updated 3 years ago
How to finetune the model (generated by train.py) with dense annotations (train + val)?
Updated 3 years ago
How to create features_faster_rcnn_x101_train.h5?
Closed 3 years ago
generative decoder
Closed 4 years ago
Bounding box coordinates
Updated 4 years ago1
Training step is too slow
Updated 4 years ago6
Torch=1.0.0 is not found
Updated 4 years ago
About concat_history in dataset.py
Closed 4 years ago3
Shared memory issues with parallelization
Updated 4 years ago5
Late Fusion demo not working
Closed 4 years ago2
How to execute visual dialog demo through colab
Updated 4 years ago
Dense annotations of part of validation and training splits
Closed 5 years ago1
Closed
Closed 5 years ago
The 'answer' would be 0 if the answer is one word
Closed 5 years ago2
About image features
Closed 5 years ago13
Need suggestion about embeddings
Closed 5 years ago4
Tokenizing is slow
Closed 5 years ago8
generative decoder
Closed 5 years ago3
Extracting Actual Images
Updated 5 years ago1
extract image features
Closed 5 years ago2
visdial_1.0_val_dense_annotations.json
Closed 5 years ago1
RuntimeError: DataLoader worker (pid 22114) is killed by signal: Killed.
Closed 5 years ago2
missing file 'data/visdial_1.0_train.json' when running train.py
Closed 5 years ago3
How to get multi-gpu to work?
Closed 5 years ago4
ffi.lua:56 expected align(#) on line 579
Closed 5 years ago7
VisDial v0.9
Closed 6 years ago2
Softmax dimension is wrong.
Closed 6 years ago1