Issues
- 0
Reported Train-Loss at epoch 200
#23 opened by CSLR-research - 0
full checkpoint
#22 opened by LiewFeng - 0
Why is the visible patch simply concatenated with the invisible mask in the decoder, rather than being placed at specific positions? There seems to be a problem with this section of the code.
#21 opened by thibault-wch - 0
- 0
请问能否提供一下Swin-T UM-MAE预训练阶段的log?感谢!
#19 opened by zengshao0622 - 0
May be a bug?
#18 opened by RachelTeamo - 4
Hello,may i ask you some questions
#2 opened by wudizuixiaosa - 0
Will you train the pvt-large in the future?
#17 opened by Roger-Liang - 1
- 0
- 0
- 0
- 1
CUDA error for Swin models
#9 opened by Haochen-Wang409 - 2
- 2
- 1
Semantic segmentation Pretrained model?
#8 opened by wuyuanmm - 2
Question about the Uniform Sampling
#7 opened by Andy1621 - 2
Can't use provided checkpoint to pre-train
#6 opened by Asers387 - 2
- 2
- 1
Can you share also HTC++ configs?
#4 opened by ckddls1321 - 1