ViTAE-Transformer/ViTDet
Unofficial implementation for [ECCV'22] "Exploring Plain Vision Transformer Backbones for Object Detection"
PythonApache-2.0
Issues
- 1
Runtime failure with customize dataset
#27 opened by robinren03 - 1
Adjusting image size
#24 opened by austinmw - 3
size mismatch and AssertionError
#22 opened by xmdgaoxin - 1
Why do I adjust my img_size to (400,600) and report the following error? QAQ
#12 opened by fadaishaitaiyang - 0
Welcome update to OpenMMLab 2.0
#26 opened by vansin - 0
when use norm_cfg, there is an error
#25 opened by mxy0610 - 0
How does mmdet training on TPUs ?
#23 opened by AkideLiu - 5
- 0
[Question] MAE vitdet interval
#21 opened by chagmgang - 0
- 2
- 2
- 2
"num_classes" error when training customize dataset
#17 opened by foolhard - 1
master_port is already used
#16 opened by YIFanH - 2
- 2
Does the VitDet still adopt FPN rather than the simple pyramid structure in VITDet paper?
#14 opened by larenzhang - 0
The model converges slowly during training
#13 opened by xiyufeng2 - 2
input must be contiguous
#11 opened by yitianlian - 5
Input shape error
#10 opened by yitianlian - 3
INDEXERROR
#8 opened by rookie0414 - 1
- 12
Getting RuntimeError
#3 opened by snailrowen1337 - 2
Environment Setting
#6 opened by junchen14 - 4
Details about MAE pre-trained ViTAE
#4 opened by Yuxin-CV - 2
Question of PCM
#5 opened by HerrYu123 - 2
How long does it take to train?
#1 opened by eshoyuan - 2
about lsj_100e config
#2 opened by youngwanLEE