Cadene vqa.pytorch issues

Cadene / vqa.pytorch

Visual Question Answering in Pytorch

716 stars 178 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

default vqa2/mutan_att_train example (also for demo) is messed

#54 dearkafka opened 4 years ago
0
What should be the size of the input_question in engine.py ?

#53 jiteshm17 opened 4 years ago
2
sysntax error in line 80

#52 ghost opened 4 years ago
1
How did you create the attention maps in your readme?

#51 Karol-G closed 4 years ago
7
Updated links

#50 jancen0 closed 4 years ago
0
FileNotFoundError: [Errno 2] No such file or directory: 'data/vqa2/raw/annotations/mscoco_train2014_annotations.json'

#49 jancen0 closed 4 years ago
1
Noticed a typo

#48 jancen0 closed 4 years ago
0
THCudaCheck FAIL file=/opt/conda/conda-bld/pytorch_1513368888240/work/torch/lib/THC/generic/THCStorage.c line=82 error=2 : out of memory

#47 huy99ls01 opened 5 years ago
0
Feature extraction: CUDA out of memory

#46 AgarwalVedika opened 5 years ago
1
How to process the multiple choice answer

#45 WangWenshan opened 5 years ago
1
LSTM avaliable?

#44 XiaosongLiXxs opened 6 years ago
0
Results on test split

#43 zwx8981 opened 6 years ago
0
The Link doesn't work: MutanNoAtt trained on the VQA 1.0 trainset

#42 haniomar closed 6 years ago
1
Why don't you apply a softmax function before the final prediction?

#41 zwx8981 closed 6 years ago
2
Batch first in LSTM

#40 DenisDsh opened 6 years ago
0
Left padding when encoding questions

#39 DenisDsh opened 6 years ago
0
'OpenEnded_mscoco_val2014_model_accuracy.json' is not created

#38 zizhu1234 closed 6 years ago
2
No learning rate decay?

#37 jakezhaojb opened 6 years ago
1
Hi, Cadene, why don't you update the feature extractor when training?

#36 zengxianyu opened 6 years ago
1
iteration over a 0-d tensor when trying to run the demo

#35 marcotcr closed 6 years ago
7
Hi cadene,why i get this error when i use demo?

#34 Rechardgu opened 6 years ago
1
Reproduce the problem using fusion scheme of concatenation

#33 vuhoangminh closed 6 years ago
2
why do i get this error on the demo server?

#32 youngjaean opened 6 years ago
0
Hi Cadene, is the Tucker decomposition part in fusion.py?

#31 ghost opened 6 years ago
1
I do not have a demo server

#30 youngjaean closed 6 years ago
2
Extracting features from Visual Genome gives error

#29 mrfarazi closed 6 years ago
2
Can we directly train the model on VisualGenome?

#28 ghost opened 6 years ago
1
Fix minor typo

#27 JustinShenk closed 6 years ago
1
Can you tell me what accuracy have you achieved?

#26 greathope opened 6 years ago
4
File "/home/vblab/Downloads/vqa.pytorch-master/vqa/datasets/features.py", line 65, in get_by_name\n index = self.name_to_index[image_name]\nKeyError: \'COCO_train2014_000000111124.jpg\'\n'

#25 jijibn opened 6 years ago
1
Have you met this question.

#24 bjtuweb-12283027 closed 7 years ago
0
Make training commands work

#23 ssnl closed 7 years ago
1
coco.py uses non-existent method os.execute

#22 ssnl closed 7 years ago
4
The results is trained on trainval + vegenome for MUTAN ATT Model ?

#21 gaopeng-eugene closed 7 years ago
1
loss decreasing is very slow

#20 eustcPL closed 7 years ago
8
Will you share code for single image inference?

#19 gaopeng-eugene closed 7 years ago
1
model specifications not coherent with the MLB paper

#18 backpropper opened 7 years ago
3
Have you ever try l2 normalization in your experiment?

#17 gaopeng-eugene closed 7 years ago
1
can you report a single model MUTAN + Att performance?

#16 gaopeng-eugene closed 7 years ago
2
Will you share your code on emsemble model

#15 gaopeng-eugene closed 7 years ago
2
How to train on VQA 2?

#14 gaopeng-eugene closed 7 years ago
0
Information regarding training time

#13 backpropper opened 7 years ago
6
MLBNoAtt does not use BayesianGRU

#12 backpropper closed 7 years ago
3
correct coco error

#11 backpropper closed 7 years ago
1
extract.py gives error for COCOImages

#10 backpropper closed 7 years ago
2
There are three ways to preprocess the question, what is the difference ?

#9 gaopeng-eugene closed 7 years ago
1
training MUTAN+Att using pytorch code achieve low accuracy

#8 gaopeng-eugene closed 7 years ago
8
Performance of MLB

#7 gaopeng-eugene closed 7 years ago
3
Training one model on multiple GPUs simultaneously not working (possible deadlock?)

#6 ahmedmagdiosman closed 7 years ago
12
Extracting image features from Visual Genome fails with multi-gpu

#5 ahmedmagdiosman closed 7 years ago
2