aimagelab meshed-memory-transformer issues

aimagelab / meshed-memory-transformer

Meshed-Memory Transformer for Image Captioning. CVPR 2020

BSD 3-Clause "New" or "Revised" License

517 stars 136 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

I can't download coco_detectons.hdf5

#48 zhu-xinru opened 3 years ago
3
Unable to open file (file read failed)

#47 batooooo closed 3 years ago
0
the caption is incomplete

#46 whqwill opened 3 years ago
22
Error on testing the network on Windows 10

#45 marcomameli1992 opened 3 years ago
6
How to inference video with this model?

#44 duanqipeng opened 3 years ago
0
The detection features cannot be download. Where can I get the detection features?

#43 UniBester opened 3 years ago
1
register_state or register_buffer ?

#42 xuewyang opened 3 years ago
1
Cannot reproduce the results during RL training stage

#41 XuMengyaAmy opened 3 years ago
1
About the data for COCO online test server

#40 Cranooooooo closed 3 years ago
3
variance dtype issue

#39 Bueno1887 opened 3 years ago
7
Meshed decoder for 6 layers Transformer

#38 HN123-123 closed 3 years ago
1
Scripts about "out-of-domain" captioning

#37 XuMengyaAmy opened 3 years ago
0
Architecture related doubt

#36 shreyanshchordia closed 3 years ago
2
Reproduce results with test.py

#35 XuMengyaAmy closed 3 years ago
1
About reproduce CIDer value?

#34 66xiaowei opened 4 years ago
3
features (N, 2048) = graph node features? & How to switch "mode='teacher_forcing" to mode='feedback' when testing

#33 XuMengyaAmy closed 3 years ago
1
Performance without reinforce optimization stage

#32 wdyin opened 4 years ago
1
Learned a priori knowledge & New dataset which is very different from MSCOCO

#31 XuMengyaAmy closed 4 years ago
2
about HDF5 file boxes shape ？

#30 BetterZH closed 3 years ago
1
Can captions be generated for new images out from the coco dataset

#29 vinod-13 opened 4 years ago
6
The input tensor shape of the self-attention in decoder when doing the beam search

#28 jlxy closed 4 years ago
1
Should mask for padding be used on the attention weights?

#27 jlxy closed 4 years ago
1
Is this code using the karpathy splits?

#26 jlxy closed 4 years ago
1
The purpose for saveing different states during beam search

#25 junchen14 closed 4 years ago
0
the purpose for different regestered states

#24 junchen14 closed 4 years ago
0
module cupy has no attribute rawkernel

#23 2014gaokao closed 4 years ago
0
Beamsearch

#22 y78h11b09 closed 4 years ago
2
visualize the attention part like results.png

#21 jkllbn2563 closed 4 years ago
4
Feature for COCO on-line Test images

#20 Jian-Xi closed 4 years ago
5
Conda Env Create Failed

#19 Mollylulu opened 4 years ago
5
Reg. Training time

#18 deepak242424 closed 4 years ago
1
Information leakage? Decoder is taking in the full output sequence during evaluation?

#17 GavinGuan95 closed 4 years ago
3
Confirming Training Time/Memory Information

#16 lukemelas closed 4 years ago
2
Q/A visual for coding

#15 TrungThanhTran opened 4 years ago
2
Flag for cpu only evalutaion.

#14 Akashtyagi closed 4 years ago
4
Random output after several early epoch then start training

#13 TrungThanhTran opened 4 years ago
0
Unable to open file (file signature not found)

#12 bean-point closed 4 years ago
1
The model runs quite slowly

#11 TrungThanhTran opened 4 years ago
1
Parallelizing the Network

#10 mujtabaasif opened 4 years ago
0
Unable to download Coco feature files.

#9 Akashtyagi closed 4 years ago
5
I meet error but i can't solve,please help

#8 bean-point closed 4 years ago
1
From features of new images to M2 transformer

#7 alesolano closed 4 years ago
3
about batch grouping

#6 homelifes closed 4 years ago
0
Unable to replicate results after retraining

#5 aemrey closed 4 years ago
5
questions regarding the paper

#4 fawazsammani closed 3 years ago
3
evaluation error

#3 lixiangpengcs closed 4 years ago
4
Is it possible to run this on a new input image?

#2 nelaturuharsha closed 4 years ago
2
Create LICENSE

#1 marcellacornia closed 4 years ago
0